Toward an ecoregion scale evaluation of eDNA metabarcoding primers: A case study for the freshwater fish biodiversity of the Murray–Darling Basin (Australia)

High‐throughput sequencing of environmental DNA (i.e., eDNA metabarcoding) has become an increasingly popular method for monitoring aquatic biodiversity. At present, such analyses require target‐specific primers to amplify DNA barcodes from co‐occurring species, and this initial amplification can introduce biases. Understanding the performance of different primers is thus recommended prior to undertaking any metabarcoding initiative. While multiple software programs are available to evaluate metabarcoding primers, all programs have their own strengths and weaknesses. Therefore, a robust in silico workflow for the evaluation of metabarcoding primers will benefit from the use of multiple programs. Furthermore, geographic differences in species biodiversity are likely to influence the performance of metabarcoding primers and further complicate the evaluation process. Here, an in silico workflow is presented that can be used to evaluate the performance of metabarcoding primers on an ecoregion scale. This workflow was used to evaluate the performance of published and newly developed eDNA metabarcoding primers for the freshwater fish biodiversity of the Murray–Darling Basin (Australia). To validate the in silico workflow, a subset of the primers, including one newly designed primer pair, were used in metabarcoding analyses of an artificial DNA community and eDNA samples. The results show that the in silico workflow allows for a robust evaluation of metabarcoding primers and can reveal important trade‐offs that need to be considered when selecting the most suitable primer. Additionally, a new primer pair was described and validated that allows for more robust taxonomic assignments and is less influenced by primer biases compared to commonly used fish metabarcoding primers.     

Genetic monitoring of open ocean biodiversity: An evaluation of DNA metabarcoding for processing continuous plankton recorder samples

DNA metabarcoding is an efficient method for measuring biodiversity, but the process of initiating long‐term DNA‐based monitoring programmes, or integrating with conventional programs, is only starting. In marine ecosystems, plankton surveys using the continuous plankton recorder (CPR) have characterized biodiversity along transects covering millions of kilometres with time‐series spanning decades. We investigated the potential for use of metabarcoding in CPR surveys. Samples (n = 53) were collected in two Southern Ocean transects and metazoans identified using standard microscopic methods and by high‐throughput sequencing of a cytochrome c oxidase subunit I marker. DNA increased the number of metazoan species identified and provided high‐resolution taxonomy of groups problematic in conventional surveys (e.g., larval echinoderms and hydrozoans). Metabarcoding also generally produced more detections than microscopy, but this sensitivity may make cross‐contamination during sampling a problem. In some samples, the prevalence of DNA from large plankton such as krill masked the presence of smaller species. We investigated adding a fixed amount of exogenous DNA to samples as an internal control to allow determination of relative plankton biomass. Overall, the metabarcoding data represent a substantial shift in perspective, making direct integration into current long‐term time‐series challenging. We discuss a number of hurdles that exist for progressing DNA metabarcoding from the current snapshot studies to the requirements of a long‐term monitoring programme. Given the power and continually increasing efficiency of metabarcoding, it is almost certain this approach will play an important role in future plankton monitoring.     

A framework for estimating the sensitivity of eDNA surveys

Imperfect sensitivity, or imperfect detection, is a feature of all survey methods that needs to be accounted for when interpreting survey results. Detection of environmental DNA (eDNA) is increasingly being used to infer species distributions, yet the sensitivity of the technique has not been fully evaluated. Sensitivity, or the probability of detecting target DNA given it is present at a site, will depend on both the survey method and the concentration and dispersion of target DNA molecules at a site. We present a model to estimate target DNA concentration and dispersion at survey sites and to estimate the sensitivity of an eDNA survey method. We fitted this model to data from a species‐specific eDNA survey for Oriental weatherloach, Misgurnus anguillicaudatus, at three sites sampled in both autumn and spring. The concentration of target DNA molecules was similar at all three sites in autumn but much higher at two sites in spring. Our analysis showed the survey method had ≥95% sensitivity at sites where target DNA concentrations were ≥11 molecules per litre. We show how these data can be used to compare sampling schemes that differ in the number of field samples collected per site and number of PCR replicates per sample to achieve ≥95% sensitivity at a given target DNA concentration. These models allow researchers to quantify the sensitivity of eDNA survey methods to optimize the probability of detecting target species, and to compare DNA concentrations spatially and temporarily.     

In silico and empirical evaluation of twelve metabarcoding primer sets for insectivorous diet analyses

During the most recent decade, environmental DNA metabarcoding approaches have been both developed and improved to minimize the biological and technical biases in these protocols. However, challenges remain, notably those relating to primer design. In the current study, we comprehensively assessed the performance of ten COI and two 16S primer pairs for eDNA metabarcoding, including novel and previously published primers. We used a combined approach of in silico, in vivo‐mock community (33 arthropod taxa from 16 orders), and guano‐based analyses to identify primer sets that would maximize arthropod detection and taxonomic identification, successfully identify the predator (bat) species, and minimize the time and financial costs of the experiment. We focused on two insectivorous bat species that live together in mixed colonies: the greater horseshoe bat (Rhinolophus ferrumequinum) and Geoffroy's bat (Myotis emarginatus). We found that primer degeneracy is the main factor that influences arthropod detection in silico and mock community analyses, while amplicon length is critical for the detection of arthropods from degraded DNA samples. Our guano‐based results highlight the importance of detecting and identifying both predator and prey, as guano samples can be contaminated by other insectivorous species. Moreover, we demonstrate that amplifying bat DNA does not reduce the primers' capacity to detect arthropods. We therefore recommend the simultaneous identification of predator and prey. Finally, our results suggest that up to one‐third of prey occurrences may be unreliable and are probably not of primary interest in diet studies, which may decrease the relevance of combining several primer sets instead of using a single efficient one. In conclusion, this study provides a pragmatic framework for eDNA primer selection with respect to scientific and methodological constraints.     

DNA metabarcoding reveals that 200‐μm‐size‐fractionated filtering is unable to discriminate between planktonic microbial and large eukaryotes

Microeukaryotic plankton (0.2–200 μm) are critical components of aquatic ecosystems and key players in global ecological processes. High‐throughput sequencing is currently revolutionizing their study on an unprecedented scale. However, it is currently unclear whether we can accurately, effectively and quantitatively depict the microeukaryotic plankton communities using traditional size‐fractionated filtering combined with molecular methods. To address this, we analysed the eukaryotic plankton communities both with, and without, prefiltering with a 200 μm pore‐size sieve –by using SSU rDNA‐based high‐throughput sequencing on 16 samples with three replicates in each sample from two subtropical reservoirs sampled from January to October in 2013. We found that ~25% reads were classified as metazoan in both size groups. The species richness, alpha and beta diversity of plankton community and relative abundance of reads in 99.2% eukaryotic OTUs showed no significant changes after prefiltering with a 200 μm pore‐size sieve. We further found that both >0.2 μm and 0.2–200 μm eukaryotic plankton communities, especially the abundant plankton subcommunities, exhibited very similar, and synchronous, spatiotemporal patterns and processes associated with almost identical environmental drivers. The lack of an effect on community structure from prefiltering suggests that environmental DNA from larger metazoa is introduced into the smaller size class. Therefore, size‐fractionated filtering with 200 μm is insufficient to discriminate between the eukaryotic plankton size groups in metabarcoding approaches. Our results also highlight the importance of sequencing depth, and strict quality filtering of reads, when designing studies to characterize microeukaryotic plankton communities.     

Helping decision making for reliable and cost‐effective 2b‐RAD sequencing and genotyping analyses in non‐model species

High‐throughput sequencing has revolutionized population and conservation genetics. RAD sequencing methods, such as 2b‐RAD, can be used on species lacking a reference genome. However, transferring protocols across taxa can potentially lead to poor results. We tested two different IIB enzymes (AlfI and CspCI) on two species with different genome sizes (the loggerhead turtle Caretta caretta and the sharpsnout seabream Diplodus puntazzo) to build a set of guidelines to improve 2b‐RAD protocols on non‐model organisms while optimising costs. Good results were obtained even with degraded samples, showing the value of 2b‐RAD in studies with poor DNA quality. However, library quality was found to be a critical parameter on the number of reads and loci obtained for genotyping. Resampling analyses with different number of reads per individual showed a trade‐off between number of loci and number of reads per sample. The resulting accumulation curves can be used as a tool to calculate the number of sequences per individual needed to reach a mean depth ≥20 reads to acquire good genotyping results. Finally, we demonstrated that selective‐base ligation does not affect genomic differentiation between individuals, indicating that this technique can be used in species with large genome sizes to adjust the number of loci to the study scope, to reduce sequencing costs and to maintain suitable sequencing depth for a reliable genotyping without compromising the results. Here, we provide a set of guidelines to improve 2b‐RAD protocols on non‐model organisms with different genome sizes, helping decision‐making for a reliable and cost‐effective genotyping.     

Detection and persistence of environmental DNA from an invasive,terrestrial mammal

Invasive Sus scrofa, a species commonly referred to as wild pig or feral swine, is a destructive invasive species with a rapidly expanding distribution across the United States. We used artificial wallows and small waterers to determine the minimum amount of time needed for pig eDNA to accumulate in the water source to a detectable level. We removed water from the artificial wallows and tested eDNA detection over the course of 2 weeks to understand eDNA persistence. We show that our method is sensitive enough to detect very low quantities of eDNA shed by a terrestrial mammal that has limited interaction with water. Our experiments suggest that the number of individuals shedding into a water system can affect persistence of eDNA. Use of an eDNA detection technique can benefit management efforts by providing a sensitive method for finding even small numbers of individuals that may be elusive using other methods.     

Genetic variability and phylogeographical patterns of a nonindigenous species invasion: a comparison of exotic vs. native zebra and quagga mussel populations

There have been few investigations of the number of founding sources and amount of genetic variability that lead to a successful nonindigenous species invasion, although genetic diversity is believed to play a central role. In the present study, population genetic structure, diversity and divergence patterns were analysed for the zebra mussel Dreissena polymorpha [n=280 samples and 63 putative randomly amplified polymorphic DNA (RAPDs) gene loci] and the quagga mussel D. bugensis (n=136 and 52 loci) from 10 nonindigenous North American and six Eurasian sampling sites, representing their present‐day ranges. Results showed that exotic populations of zebra and quagga mussels had surprisingly high genetic variability, similar to those in the Eurasian populations, suggesting large numbers of founding individuals and consistent with the hypothesis of multiple colonizations. Patterns of genetic relationships indicate that the North American populations of D. polymorpha likely were founded by multiple source populations from north‐western and northcentral Europe, but not from southcentral or eastern Europe. Sampling areas within North America also were significantly divergent, having levels of gene flow and migration about twice those separating long‐established Eurasian populations. Samples of D. bugensis in Lakes Erie and Ontario were significantly different, with the former being more closely related to a native population from the Dnieper River, Ukraine. No evidence for a founder effect was discerned for either species.     

Methodological considerations for detection of terrestrial small‐body salamander eDNA and implications for biodiversity conservation

Environmental DNA (eDNA) can be used as an assessment tool to detect populations of threatened species and provide fine‐scale data required to make management decisions. The objectives of this project were to use quantitative PCR (qPCR) to: (i) detect spiked salamander DNA in soil, (ii) quantify eDNA degradation over time, (iii) determine detectability of salamander eDNA in a terrestrial environment using soil, faeces, and skin swabs, (iv) detect salamander eDNA in a mesocosm experiment. Salamander eDNA was positively detected in 100% of skin swabs and 66% of faecal samples and concentrations did not differ between the two sources. However, eDNA was not detected in soil samples collected from directly underneath wild‐caught living salamanders. Salamander genomic DNA (gDNA) was detected in all qPCR reactions when spiked into soil at 10.0, 5.0, and 1.0 ng/g soil and spike concentration had a significant effect on detected concentrations. Only 33% of samples showed recoverable eDNA when spiked with 0.25 ng/g soil, which was the low end of eDNA detection. To determine the rate of eDNA degradation, gDNA (1 ng/g soil) was spiked into soil and quantified over seven days. Salamander eDNA concentrations decreased across days, but eDNA was still amplifiable at day 7. Salamander eDNA was detected in two of 182 mesocosm soil samples over 12 weeks (n = 52 control samples; n = 65 presence samples; n = 65 eviction samples). The discrepancy in detection success between experiments indicates the potential challenges for this method to be used as a monitoring technique for small‐bodied wild terrestrial salamander populations.     

Identification of Belgian mosquito species (Diptera: Culicidae) by DNA barcoding

Since its introduction in 2003, DNA barcoding has proven to be a promising method for the identification of many taxa, including mosquitoes (Diptera: Culicidae). Many mosquito species are potential vectors of pathogens, and correct identification in all life stages is essential for effective mosquito monitoring and control. To use DNA barcoding for species identification, a reliable and comprehensive reference database of verified DNA sequences is required. Hence, DNA sequence diversity of mosquitoes in Belgium was assessed using a 658 bp fragment of the mitochondrial cytochrome oxidase I (COI) gene, and a reference data set was established. Most species appeared as well‐supported clusters. Intraspecific Kimura 2‐parameter (K2P) distances averaged 0.7%, and the maximum observed K2P distance was 6.2% for Aedes koreicus. A small overlap between intra‐ and interspecific K2P distances for congeneric sequences was observed. Overall, the identification success using best match and the best close match criteria were high, that is above 98%. No clear genetic division was found between the closely related species Aedes annulipes and Aedes cantans, which can be confused using morphological identification only. The members of the Anopheles maculipennis complex, that is Anopheles maculipennis s.s. and An. messeae, were weakly supported as monophyletic taxa. This study showed that DNA barcoding offers a reliable framework for mosquito species identification in Belgium except for some closely related species.     

Toward accurate molecular identification of species in complex environmental samples: testing the performance of sequence filtering and clustering methods

Metabarcoding has the potential to become a rapid, sensitive, and effective approach for identifying species in complex environmental samples. Accurate molecular identification of species depends on the ability to generate operational taxonomic units (OTUs) that correspond to biological species. Due to the sometimes enormous estimates of biodiversity using this method, there is a great need to test the efficacy of data analysis methods used to derive OTUs. Here, we evaluate the performance of various methods for clustering length variable 18S amplicons from complex samples into OTUs using a mock community and a natural community of zooplankton species. We compare analytic procedures consisting of a combination of (1) stringent and relaxed data filtering, (2) singleton sequences included and removed, (3) three commonly used clustering algorithms (mothur, UCLUST, and UPARSE), and (4) three methods of treating alignment gaps when calculating sequence divergence. Depending on the combination of methods used, the number of OTUs varied by nearly two orders of magnitude for the mock community (60–5068 OTUs) and three orders of magnitude for the natural community (22–22191 OTUs). The use of relaxed filtering and the inclusion of singletons greatly inflated OTU numbers without increasing the ability to recover species. Our results also suggest that the method used to treat gaps when calculating sequence divergence can have a great impact on the number of OTUs. Our findings are particularly relevant to studies that cover taxonomically diverse species and employ markers such as rRNA genes in which length variation is extensive.     

Invasion genetics of American cherry fruit fly in Europe and signals of hybridization with the European cherry fruit fly

The American cherry fruit fly is an invasive pest species in Europe, of serious concern in tart cherry production as well as for the potential to hybridize with the European cherry fruit fly, Rhagoletis cerasi L. (Diptera: Tephritidae), which might induce new pest dynamics. In the first European reports, the question arose whether only the eastern American cherry fruit fly, Rhagoletis cingulata (Loew) (Diptera: Tephritidae), is present, or also the closely related western American cherry fruit fly, Rhagoletis indifferens Curran. In this study, we investigate the species status of European populations by comparing these with populations of both American species from their native ranges, the invasion dynamics in German (first report in 1993) and Hungarian (first report in 2006) populations, and we test for signals of hybridization with the European cherry fruit fly. Although mtDNA sequence genealogy could not separate the two American species, cross‐species amplification of 14 microsatellite loci separated them with high probabilities (0.99–1.0) and provided evidence for R. cingulata in Europe. German and Hungarian R. cingulata populations differed significantly in microsatellite allele frequencies, mtDNA haplotype and wing pattern distributions, and both were genetically depauperate relative to North American populations. The diversity suggests independent founding events in Germany and Hungary. Within each country, R. cingulata displayed little or no structure in any trait, which agrees with rapid local range expansions. In cross‐species amplifications, signals of hybridization between R. cerasi and R. cingulata were found in 2% of R. cingulata individuals and in 3% of R. cerasi. All putative hybrids had R. cerasi mtDNA indicating that the original between‐species mating involved R. cerasi females and R. cingulata males.     

Ecological specialization and niche overlap of subterranean rodents inferred from DNA metabarcoding diet analysis

Knowledge of how animal species use food resources available in the environment can increase our understanding of many ecological processes. However, obtaining this information using traditional methods is difficult for species feeding on a large variety of food items in highly diverse environments. We amplified the DNA of plants for 306 scat and 40 soil samples, and applied an environmental DNA metabarcoding approach to investigate food preferences, degree of diet specialization and diet overlap of seven herbivore rodent species of the genus Ctenomys distributed in southern and midwestern Brazil. The metabarcoding approach revealed that these species consume more than 60% of the plant families recovered in soil samples, indicating generalist feeding habits of ctenomyids. The family Poaceae was the most common food resource retrieved in scats of all species as well in soil samples. Niche overlap analysis indicated high overlap in the plant families and molecular operational taxonomic units consumed, mainly among the southern species. Interspecific differences in diet composition were influenced, among other factors, by the availability of resources in the environment. In addition, our results provide support for the hypothesis that the allopatric distributions of ctenomyids allow them to exploit the same range of resources when available, possibly because of the absence of interspecific competition.     

Environmental DNA detection of rare and invasive fish species in two Great Lakes tributaries

The extraction and characterization of DNA from aquatic environmental samples offers an alternative, noninvasive approach for the detection of rare species. Environmental DNA, coupled with PCR and next‐generation sequencing (“metabarcoding”), has proven to be very sensitive for the detection of rare aquatic species. Our study used a custom‐designed group‐specific primer set and next‐generation sequencing for the detection of three species at risk (Eastern Sand Darter, Ammocrypta pellucida; Northern Madtom, Noturus stigmosus; and Silver Shiner, Notropis photogenis), one invasive species (Round Goby, Neogobius melanostomus) and an additional 78 native species from two large Great Lakes tributary rivers in southern Ontario, Canada: the Grand River and the Sydenham River. Of 82 fish species detected in both rivers using capture‐based and eDNA methods, our eDNA method detected 86.2% and 72.0% of the fish species in the Grand River and the Sydenham River, respectively, which included our four target species. Our analyses also identified significant positive and negative species co‐occurrence patterns between our target species and other identified species. Our results demonstrate that eDNA metabarcoding that targets the fish community as well as individual species of interest provides a better understanding of factors affecting the target species spatial distribution in an ecosystem than possible with only target species data. Additionally, eDNA is easily implemented as an initial survey tool, or alongside capture‐based methods, for improved mapping of species distribution patterns.     

FuzzyID2: A software package for large data set species identification via barcoding and metabarcoding using hidden Markov models and fuzzy set methods

Species identification through DNA barcoding or metabarcoding has become a key approach for biodiversity evaluation and ecological studies. However, the rapid accumulation of barcoding data has created some difficulties: for instance, global enquiries to a large reference library can take a very long time. We here devise a two‐step searching strategy to speed identification procedures of such queries. This firstly uses a Hidden Markov Model (HMM) algorithm to narrow the searching scope to genus level and then determines the corresponding species using minimum genetic distance. Moreover, using a fuzzy membership function, our approach also estimates the credibility of assignment results for each query. To perform this task, we developed a new software pipeline, FuzzyID2, using Python and C++. Performance of the new method was assessed using eight empirical data sets ranging from 70 to 234,535 barcodes. Five data sets (four animal, one plant) deployed the conventional barcode approach, one used metabarcodes, and two were eDNA‐based. The results showed mean accuracies of generic and species identification of 98.60% (with a minimum of 95.00% and a maximum of 100.00%) and 94.17% (with a range of 84.40%–100.00%), respectively. Tests with simulated NGS sequences based on realistic eDNA and metabarcode data demonstrated that FuzzyID2 achieved a significantly higher identification success rate than the commonly used Blast method, and the TIPP method tends to find many fewer species than either FuzztID2 or Blast. Furthermore, data sets with tens of thousands of barcodes need only a few seconds for each query assignment using FuzzyID2. Our approach provides an efficient and accurate species identification protocol for biodiversity‐related projects with large DNA sequence data sets.