Evaluation of detection probabilities at the water-filtering and initial PCR steps in environmental DNA metabarcoding using a multispecies site occupancy model

Hideyuki Doi,Keiichi Sato,Keiichi Fukaya,Shin-Ichiro Oka,Masaki Miya,Michio Kondoh

doi:10.1038/s41598-019-40233-1

Abstract

Environmental DNA (eDNA) metabarcoding is a recently developed method to assess biodiversity based on a high-throughput parallel DNA sequencing applied to DNA present in the ecosystem. Although eDNA metabarcoding enables a rapid assessment of biodiversity, it is prone to species detection errors that may occur at sequential steps in field sampling, laboratory experiments, and bioinformatics. In this study, we illustrate how the error rates in the eDNA metabarcoding-based species detection can be accounted for by applying the multispecies occupancy modelling framework. We report a case study with the eDNA sample from an aquarium tank in which the detection probabilities of species in the two major steps of eDNA metabarcoding, filtration and PCR, across a range of PCR annealing temperatures, were examined. We also show that the results can be used to examine the efficiency of species detection under a given experimental design and setting, in terms of the efficiency of species detection, highlighting the usefulness of the multispecies site occupancy modelling framework to study the optimum conditions for molecular experiments.

Highlights

Environmental DNA methods have been increasingly considered as useful tools in the investigation of the distribution of aquatic and terrestrial macroorganisms inhabiting various habitats[1,2,3,4,5,6,7,8,9,10,11,12,13,14]
We present the results of Environmental DNA (eDNA) metabarcoding for the fish community in a large aquarium with known fish species, where replicates were taken in the filtration and the PCR (1st PCR for library preparation) steps of the laboratory experiment
The results highlight the advantage of using the multispecies occupancy modelling framework for eDNA metabarcoding, which can help to determine the number of replicates at different experimental steps, as well as estimate the efficiency of species detection in a given experiment

Summary

Introduction

Environmental DNA (eDNA) methods have been increasingly considered as useful tools in the investigation of the distribution of aquatic and terrestrial macroorganisms inhabiting various habitats[1,2,3,4,5,6,7,8,9,10,11,12,13,14]. Ad hoc procedures have been proposed, such as not considering species detected in just a few PCR replicates Despite these methodological issues, which are obviously critical to the assessment of biodiversity based on eDNA metabarcoding, they have not yet been well investigated, especially for laboratory experiments. The objective of this study was to illustrate how the multispecies occupancy modelling framework can be used to evaluate probabilities of species detection in different steps of laboratory experiments for eDNA metabarcoding. The effect of the PCR annealing temperature has been shown to affect DNA metabarcoding and the use of inappropriate PCR conditions can affect the final taxonomic assignment in metazoan metabarcoding analyses[34] Given these estimates of detection probabilities, we show that the effectiveness of an experimental design and setting can be evaluated in terms of the efficiency of species detection

Objectives

Methods

Results

Conclusion