Search

The accurate extraction of species-abundance information from DNA-based data (metabarcoding, metagenomics) could contribute usefully to diet analysis and food-web reconstruction, the inference of species interactions, the modelling of population dynamics and species distributions, the biomonitoring of environmental state and change, and the inference of false positives and negatives. However, multiple sources of bias and noise in sampling and processing combine to inject error into DNA-based datasets. We focus here on the laboratory and bioinformatic processes of generating DNA-based data, since sampling bias and noise are addressed extensively in the ecological literature. To extract abundance information, it is useful to distinguish two concepts. (1)Within-sample across-speciesquantification describes relative species abundances within one sample. (2)Across-sample within-speciesquantification describes how the abundance of each individual species varies from sample to sample, as in a time series, an environmental gradient, or experimental treatments. First, we review the literature on methods to recover (1)across-speciesabundance information (which is achieved by removing what we call ‘species pipeline biases’) and (2)within-speciesabundance information (by removing what we call ‘pipeline noise’). We argue that many ecological questions can be answered by extracting only within-species quantification, and we therefore demonstrate how to use a ‘DNA spike-in’ to correct for pipeline noise and recover within-speciesabundance information. We also introduce a model-based estimator that can be employed on datasets without a physical spike-in to approximately estimate and correct for pipeline noise.

Species interactions are known to structure ecological communities. Still, the influence of climate change on biodiversity has primarily been evaluated by correlating individual species distributions with local climatic descriptors, then extrapolating into future climate scenarios. We ask whether predictions on arctic arthropod response to climate change can be improved by accounting for species interactions. For this, we use a 14-year-long, weekly time series from Greenland, resolved to the species level by mitogenome mapping. During the study period, temperature increased by 2 degrees C and arthropod species richness halved. We show that with abiotic variables alone, we are essentially unable to predict species responses, but with species interactions included, the predictive power of the models improves considerably. Cascading trophic effects thereby emerge as important in structuring biodiversity response to climate change. Given the need to scale up from species-level to community-level projections of biodiversity change, these results represent a major step forward for predictive ecology.

1AbstractProtected areas are central to meeting biodiversity conservation goals, but measuring their effectiveness is challenging. We address this challenge by using DNA from leech-ingested bloodmeals to estimate vertebrate occupancies across the 677 km2 Ailaoshan reserve in Yunnan, China. 163 park rangers collected 30,468 leeches from 172 patrol areas. We identified 86 vertebrate species, including amphibians, mammals, birds, and squamates. Multi-species occupancy modelling showed that species richness increased with elevation and distance to reserve edge, including the distributions of most of the large mammals (e.g. sambar, black bear, serow, tufted deer). The exceptions were the three domestic mammal species (cows, sheep, goats) and muntjak deer, which were more common at lower elevations. eDNA-estimated vertebrate occupancies are Granular, Repeatable, Auditable, Direct, Efficient, and Simple-to-understand measures that can be used to assess conservation effectiveness and thus to improve the contributions that protected areas make to achieving global biodiversity goals.

The accurate quantification of eukaryotic species abundances from bulk samples remains a key challenge for community ecology and environmental biomonitoring. We resolve this challenge by combining shotgun sequencing, mapping to reference DNA barcodes or to mitogenomes, and three correction factors: (a) a percent-coverage threshold to filter out false positives, (b) an internal-standard DNA spike-in to correct for stochasticity during sequencing, and (c) technical replicates to correct for stochasticity across sequencing runs. The SPIKEPIPE pipeline achieves a strikingly high accuracy of intraspecific abundance estimates (in terms of DNA mass) from samples of known composition (mapping to barcodes R2 = .93, mitogenomes R2 = .95) and a high repeatability across environmental-sample replicates (barcodes R2 = .94, mitogenomes R2 = .93). As proof of concept, we sequence arthropod samples from the High Arctic, systematically collected over 17 years, detecting changes in species richness, species-specific abundances, and phenology. SPIKEPIPE provides cost-efficient and reliable quantification of eukaryotic communities.

Summary A crucial step in the use of DNA markers for biodiversity surveys is the assignment of Linnaean taxonomies (species, genus, etc.) to sequence reads. This allows the use of all the information known based on the taxonomic names. Taxonomic placement of DNA barcoding sequences is inherently probabilistic because DNA sequences contain errors, because there is natural variation among sequences within a species, and because reference data bases are incomplete and can have false annotations. However, most existing bioinformatics methods for taxonomic placement either exclude uncertainty, or quantify it using metrics other than probability. In this paper we evaluate the performance of the recently proposed probabilistic taxonomic placement method PROTAX by applying it to both annotated reference sequence data as well as to unknown environmental data. Our four case studies include contrasting taxonomic groups (fungi, bacteria, mammals and insects), variation in the length and quality of the barcoding sequences (from individually Sanger-sequenced sequences to short Illumina reads), variation in the structures and sizes of the taxonomies (800–130 000 species) and variation in the completeness of the reference data bases (representing 15–100% of known species). Our results demonstrate that PROTAX yields essentially unbiased probabilities of taxonomic placement, which means its quantification of species identification uncertainty is reliable. As expected, the accuracy of taxonomic placement increases with increasing coverage of taxonomic and reference sequence data bases, and with increasing ratio of genetic variation among taxonomic levels over within taxonomic levels. We conclude that reliable species-level identification from environmental samples is still challenging and that neglecting identification uncertainty can lead to spurious inference. A key aim for future research is the completion of taxonomic and reference sequence data bases and making these two types of data compatible.

The accurate quantification of eukaryotic species abundances from bulk samples remains a key challenge for community description and environmental biomonitoring. We resolve this challenge by combining shotgun sequencing, mapping to reference DNA barcodes or to mitogenomes, and three correction factors: (1) a percent-coverage threshold to filter out false positives, (2) an internal-standard DNA spike-in to correct for stochasticity during sequencing, and (3) technical replicates to correct for stochasticity across sequencing runs. This pipeline achieves a strikingly high accuracy of intraspecific abundance estimates from samples of known composition (mapping to barcodes R2=0.93, mitogenomes R2=0.95) and a high repeatability across environmental-sample replicates (barcodes R2=0.94, mitogenomes R2=0.93). As proof of concept, we sequence arthropod samples from the High Arctic systematically collected over 17 years, detecting changes in species richness, abundance, and phenology using either barcodes or mitogenomes. SPIKEPIPE provides cost-efficient and reliable quantification of eukaryotic communities, with direct application to environmental biomonitoring.Statement of authorshipNMS has been involved in running the BioBasis sampling program for more than twenty years. TR, NMS, DWY, and OO conceived the study and its design. TH led the work in generating all the DNA samples and YJ led the work in assembling and annotating the mitogenomes for the mitochondrial genome reference database. TH led the work in generating the mock communities and bulk samples, with contributions from YJ and JW. YJ and DWY developed the molecular and bioinformatic methods. OO led the modelling of the data. TR and OO wrote the first draft of the manuscript, and all authors contributed substantially to its further improvement.Data accessibility statementShould the manuscript be accepted, the data supporting the results will be archived in an appropriate public repository (Dryad), and the data DOI will be included at the end of the article. The bioinformatic and R scripts and associated data tables will also be made available on github.com.

1. Bee populations and other pollinators face multiple, synergistically acting threats, which have led to population declines, loss of local species richness and pollination services, and extinctions. However, our understanding of the degree, distribution and causes of declines is patchy, in part due to inadequate monitoring systems, with the challenge of taxonomic identification posing a major logistical barrier. Pollinator conservation would benefit from a high-throughput identification pipeline.\ud \ud 2. We show that the metagenomic mining and resequencing of mitochondrial genomes (mitogenomics) can be applied successfully to bulk samples of wild bees. We assembled the mitogenomes of 48 UK bee species and then shotgun-sequenced total DNA extracted from 204 whole bees that had been collected in 10 pan-trap samples from farms in England and been identified morphologically to 33 species. Each sample data set was mapped\ud against the 48 reference mitogenomes.\ud \ud 3. The morphological and mitogenomic data sets were highly congruent. Out of 63 total species detections in the morphological data set, the mitogenomic data set made 59 correct detections (93�7% detection rate) and detected\ud six more species (putative false positives). Direct inspection and an analysis with species-specific primers suggested that these putative false positives were most likely due to incorrect morphological IDs. Read frequency\ud significantly predicted species biomass frequency (R2 = 24�9%). Species lists, biomass frequencies, extrapolated\ud species richness and community structure were recovered with less error than in a metabarcoding pipeline.\ud \ud 4. Mitogenomics automates the onerous task of taxonomic identification, even for cryptic species, allowing the\ud tracking of changes in species richness and istributions. A mitogenomic pipeline should thus be able to contain\ud costs, maintain consistently high-quality data over long time series, incorporate retrospective taxonomic revisions and provide an auditable evidence trail. Mitogenomic data sets also provide estimates of species counts within samples and thus have potential for tracking population trajectories.

Strong global demand for tropical timber and agricultural products has driven large-scale logging and subsequent conversion of tropical forests. Given that the majority of tropical landscapes have been or will likely be logged, the protection of biodiversity within tropical forests thus depends on whether species can persist in these economically exploited lands, and if species cannot persist, whether we can protect enough primary forest from logging and conversion. However, our knowledge of the impact of logging and conversion on biodiversity is limited to a few taxa, often sampled in different locations with complex land-use histories, hampering attempts to plan cost-effective conservation strategies and to draw conclusions across taxa. Spanning a land-use gradient of primary forest, once- and twice-logged forests, and oil palm plantations, we used traditional sampling and DNA metabarcoding to compile an extensive data set in Sabah, Malaysian Borneo for nine vertebrate and invertebrate taxa to quantify the biological impacts of logging and oil palm, develop cost-effective methods of protecting biodiversity, and examine whether there is congruence in response among taxa. Logged forests retained high species richness, including, on average, 70% of species found in primary forest. In contrast, conversion to oil palm dramatically reduces species richness, with significantly fewer primary-forest species than found on logged forest transects for seven taxa. Using a systematic conservation planning analysis, we show that efficient protection of primary-forest species is achieved with land portfolios that include a large proportion of logged-forest plots. Protecting logged forests is thus a cost-effective method of protecting an ecologically and taxonomically diverse range of species, particularly when conservation budgets are limited. Six indicator groups (birds, leaf-litter ants, beetles, aerial hymenopterans, flies, and true bugs) proved to be consistently good predictors of the response of the other taxa to logging and oil palm. Our results confidently establish the high conservation value of logged forests and the low value of oil palm. Cross-taxon congruence in responses to disturbance also suggests that the practice of focusing on key indicator taxa yields important information of general biodiversity in studies of logging and oil palm.

A crucial step in the use of DNA markers for biodiversity surveys is the assignment of Linnaean taxonomies (species, genus, etc.) to sequence reads. This allows the use of all the information known based on the taxonomic names. Taxonomic placement of DNA barcoding sequences is inherently probabilistic because DNA sequences contain errors, because there is natural variation among sequences within a species, and because reference databases are incomplete and can have false annotations. However, most existing bioinformatics methods for taxonomic placement either exclude uncertainty, or quantify it using metrics other than probability.In this paper we evaluate the performance of a recently proposed probabilistic taxonomic placement method PROTAX by applying it to both annotated reference sequence data as well as unknown environmental data. Our four case studies include contrasting taxonomic groups (fungi, bacteria, mammals, and insects), variation in the length and quality of the barcoding sequences (from individually Sanger-sequenced sequences to short Illumina reads), variation in the structures and sizes of the taxonomies (from 800 to 130 000 species), and variation in the completeness of the reference databases (representing 15% to 100% of the species).Our results demonstrate that PROTAX yields essentially unbiased assessment of probabilities of taxonomic placement, and thus that its quantification of species identification uncertainty is reliable. As expected, the accuracy of taxonomic placement increases with increasing coverage of taxonomic and reference sequence databases, and with increasing ratio of genetic variation among taxonomic levels over within taxonomic levels.Our results show that reliable species-level identification from environmental samples is still challenging, and thus neglecting identification uncertainty can lead to spurious inference. A key aim for future research is the completion and pruning of taxonomic and reference sequence databases, and making these two types of data compatible.

Summary Metabarcoding of mixed arthropod samples for biodiversity assessment has mostly been carried out on the 454 GS FLX sequencer (Roche, Branford, Connecticut, USA), due to its ability to produce long reads (≥400 bp) that are believed to allow higher taxonomic resolution. The Illumina sequencing platforms, with their much higher throughputs, could potentially reduce sequencing costs and improve sequence quality, but the associated shorter read length (typically

For support, we thank Yunnan Province (20080A001), the Chinese Academy of Sciences (0902281081, KSCX2‐YW‐Z‐1027), the National Natural Science Foundation of China (31170498), and the University of East Anglia.

Plant diversity surely determines arthropod diversity, but only moderate correlations between arthropod and plant species richness had been observed until Basset et al. (2012, Science 338: 1481-1484) finally undertook an unprecedentedly comprehensive sampling of a tropical forest and demonstrated that plant species richness could indeed accurately predict arthropod species richness. We now require a high-throughput pipeline to operationalize this result so that we can (1) test competing explanations for tropical arthropod megadiversity, (2) improve estimates of global eukaryotic species diversity, and (3) use plant and arthropod communities as efficient proxies for each other, thus improving the efficiency of conservation planning and of detecting forest degradation and recovery. We therefore applied metabarcoding to Malaise-trap samples across two tropical landscapes in China. We demonstrate that plant species richness can accurately predict arthropod (mostly insect) species richness and that plant and insect community compositions are highly correlated, even in landscapes that are large, heterogeneous, and anthropogenically modified. Finally, we review how metabarcoding makes feasible highly replicated tests of the major competing explanations for tropical megadiversity.

Strong global demand for tropical timber and agricultural products has driven large-scale logging and subsequent conversion of tropical forests. Given that the majority of tropical landscapes have been or will likely be logged, the protection of biodiversity within tropical forests thus depends on whether species can persist in these economically exploited lands, and if species cannot persist, whether we can protect enough primary forest from logging and conversion. However, our knowledge of the impact of logging and conversion on biodiversity is limited to a few taxa, often sampled in different locations with complex land-use histories, hampering attempts to plan cost-effective conservation strategies and to draw conclusions across taxa. Spanning a land-use gradient of primary forest, once- and twice-logged forests, and oil palm plantations, we used traditional sampling and DNA metabarcoding to compile an extensive data set in Sabah, Malaysian Borneo for nine vertebrate and invertebrate taxa to quantify the biological impacts of logging and oil palm, develop cost-effective methods of protecting biodiversity, and examine whether there is congruence in response among taxa. Logged forests retained high species richness, including, on average, 70% of species found in primary forest. In contrast, conversion to oil palm dramatically reduces species richness, with significantly fewer primary-forest species than found on logged forest transects for seven taxa. Using a systematic conservation planning analysis, we show that efficient protection of primary-forest species is achieved with land portfolios that include a large proportion of logged-forest plots. Protecting logged forests is thus a cost-effective method of protecting an ecologically and taxonomically diverse range of species, particularly when conservation budgets are limited. Six indicator groups (birds, leaf-litter ants, beetles, aerial hymenopterans, flies, and true bugs) proved to be consistently good predictors of the response of the other taxa to logging and oil palm. Our results confidently establish the high conservation value of logged forests and the low value of oil palm. Cross-taxon congruence in responses to disturbance also suggests that the practice of focusing on key indicator taxa yields important information of general biodiversity in studies of logging and oil palm.

To manage and conserve biodiversity, one must know what is being lost, where, and why, as well as which remedies are likely to be most effective. Metabarcoding technology can characterise the species compositions of mass samples of eukaryotes or of environmental DNA. Here, we validate metabarcoding by testing it against three high‐quality standard data sets that were collected in Malaysia (tropical), China (subtropical) and the United Kingdom (temperate) and that comprised 55,813 arthropod and bird specimens identified to species level with the expenditure of 2,505 person‐hours of taxonomic expertise. The metabarcode and standard data sets exhibit statistically correlated alpha‐ and beta‐diversities, and the two data sets produce similar policy conclusions for two conservation applications: restoration ecology and systematic conservation planning. Compared with standard biodiversity data sets, metabarcoded samples are taxonomically more comprehensive, many times quicker to produce, less reliant on taxonomic expertise and auditable by third parties, which is essential for dispute resolution., We thank Yang Yahan, Alice Wang, Vincent Moulton, David Warton and Wadud Miah for support and advice and to Ding Zhaoli for sequencing. LA, YT, AN and RK were supported by the Queensland‐Chinese Academy of Sciences (QCAS) Biotechnology Fund (GJHZ1130) and Griffith University. DPE was supported by a STEP fellowship at Princeton University. SP was supported by the Natural Environment Research Council, Forestry Commission, Norfolk Biodiversity Information Service and Suffolk Biodiversity Partnership. Additional support for DPE, PW, FAE, THL and WHH was provided by a grant from the High Meadows Foundation to DSW. YQJ, XYW and DWY were supported by Yunnan Province (20080A001), the Chinese Academy of Sciences (0902281081, KSCX2‐YW‐Z‐1027), the National Natural Science Foundation of China (31170498), the Ministry of Science and Technology of China (2012FY110800), the University of East Anglia, and the State Key Laboratory of Genetic Resources and Evolution at the Kunming Institute of Zoology.

The horizontal distribution of stems, stand density and the differentiation of tree dimensions are among the most important aspects of stand structure. An increasing complexity of stand structure is often linked to a higher number of species and to greater ecological stability. For quantification, the Structural Complexity Index (SCI) describes structural complexity by means of an area ratio of the surface that is generated by connecting the tree tops of neighbouring trees to form triangles to the surface that is covered by all triangles if projected on a flat plane. Here, we propose two ecologically relevant modifications of the SCI: The degree of mingling of tree attributes, quantified by a vector ruggedness measure, and a stem density term. We investigate how these two modifications influence index values. Data come from forest inventory field plots sampled along a disturbance gradient from heavily disturbed shrub land, through secondary regrowth to mature montane rainforest stands in Mengsong, Xishuangbanna, Yunnan, China. An application is described linking structural complexity, as described by the SCI and its modified versions, to changes in species composition of insect communities. The results of this study show that the Enhanced Structural Complexity Index (ESCI) can serve as a valuable tool for forest managers and ecologists for describing the structural complexity of forest stands and is particularly valuable for natural forests with a high degree of structural complexity. peerReviewed