Categorization of species as native or nonnative using DNA sequence signatures without a complete reference library

Abstract

New genetic diagnostic approaches have greatly aided efforts to document global biodiversity and improve biosecurity. This is especially true for organismal groups in which species diversity has been underestimated historically due to difficulties associated with sampling, the lack of clear morphological characteristics, and/or limited availability of taxonomic expertise. Among these methods, DNA sequence barcoding (also known as “DNA barcoding”) and by extension, meta‐barcoding for biological communities, has emerged as one of the most frequently utilized methods for DNA‐based species identifications. Unfortunately, the use of DNA barcoding is limited by the availability of complete reference libraries (i.e., a collection of DNA sequences from morphologically identified species), and by the fact that the vast majority of species do not have sequences present in reference databases. Such conditions are critical especially in tropical locations that are simultaneously biodiversity rich and suffer from a lack of exploration and DNA characterization by trained taxonomic specialists. To facilitate efforts to document biodiversity in regions lacking complete reference libraries, we developed a novel statistical approach that categorizes unidentified species as being either likely native or likely nonnative based solely on measures of nucleotide diversity. We demonstrate the utility of this approach by categorizing a large sample of specimens of terrestrial insects and spiders (collected as part of the Moorea BioCode project) using a generalized linear mixed model (GLMM). Using a training data set of known endemic (n = 45) and known introduced species (n = 102), we then estimated the likely native/nonnative status for 4,663 specimens representing an estimated 1,288 species (412 identified species), including both those specimens that were either unidentified or whose endemic/introduced status was uncertain. Using this approach, we were able to increase the number of categorized specimens by a factor of 4.4 (from 794 to 3,497), and the number of categorized species by a factor of 4.8 from (147 to 707) at a rate much greater than chance (77.6% accuracy). The study identifies phylogenetic signatures of both native and nonnative species and suggests several practical applications for this approach including monitoring biodiversity and facilitating biosecurity.

Comparison of markers for the monitoring of freshwater benthic biodiversity through DNA metabarcoding

“…Our study focused on European taxa, where taxonomic knowledge is particularly good (Brewer et al, 2012;Moustakas & Karakassis, 2005;Rodrigues et al, 2010) and, with targeted studies, we could envisage an improvement of database completeness in the next few years. However, our results on primer performance can be also useful in megadiverse, tropical areas, where taxonomy-free biomonitoring is a viable option (Andersen et al, 2019).…”

“…eDNA and metabarcoding are therefore extremely promising for the assessment of freshwater communities (Hering et al, 2018;Kuntke et al, 2020;Li et al, 2018). For metabarcoding, DNA can be extracted from the tissue of pooled macroinvertebrate communities, amplified using universal primers, sequenced and identified on the basis of reference databases (Andújar et al, 2018;Baird & Hajibabaei, 2012;Yu et al, 2012). This approach uses the same starting material as traditional biomonitoring, but avoids the complexity of morphology-based taxonomy (Baird & Hajibabaei, 2012).…”

Integrating eDNA metabarcoding and simultaneous underwater visual surveys to describe complex fish communities in a marine biodiversity hotspot

“…A key element in eDNA reference-based metabarcoding is that the taxonomic determination of amplicons is established by comparison with a taxonomically annotated reference database (Mendoza-Zepeda et al, 2014). Thorough and accurate detection of species from DNA amplicons requires a taxonomically comprehensive coverage of known taxa in reference databases (Andersen et al, 2019;Weigand et al, 2019). However, such reference databases are often missing for local faunas (Juhel et al, 2020) and their completeness varies for different regions.…”

Community assembly and metaphylogeography of soil biodiversity: insights from haplotype-level community DNA metabarcoding within an oceanic island

“…These findings are in agreement with Cicconardi et al (2017), who concluded from genome skimming data that 88% of the 25 Collembola species they sampled from laurel forests in Tenerife result from human-mediated introductions. Distinguishing between native and introduced origins for soil-adapted species is challenging, and focused studies are needed to elucidate the extent of species introductions within oceanic islands (Andersen et al, 2019).…”