Predicting the Ecological Quality Status of Marine Environments from eDNA Metabarcoding Data Using Supervised Machine Learning

Alignment-free classification of COI DNA barcode data with the Python package Alfie

Characterization of biodiversity from environmental DNA samples and bulk metabarcoding data is hampered by off-target sequences that can confound conclusions about a taxonomic group of interest. Existing methods for isolation of target sequences rely on alignment to existing reference barcodes, but this can bias results against novel genetic variants. Effectively parsing targeted DNA barcode data from off-target noise improves the quality of biodiversity estimates and biological conclusions by limiting subsequent analyses to a relevant subset of available data. Here, we present Alfie, a Python package for the alignment-free classification of cytochrome c oxidase subunit I (COI) DNA barcode sequences to taxonomic kingdoms. The package determines k-mer frequencies of DNA sequences, and the frequencies serve as input for a neural network classifier that was trained and tested using ~58,000 publicly available COI sequences. The classifier was designed and optimized through a series of tests that allowed for the optimal set of DNA k-mer features and optimal machine learning algorithm to be selected. The neural network classifier rapidly assigns COI sequences of varying lengths to kingdoms with greater than 99% accuracy and is shown to generalize effectively and make accurate predictions about data from previously unseen taxonomic classes. The package contains an application programming interface that allows the Alfie package’s functionality to be extended to different DNA sequence classification tasks to suit a user’s need, including classification of different genes and barcodes, and classification to different taxonomic levels. Alfie is free and publicly available through GitHub (https://github.com/CNuge/alfie) and the Python package index (https://pypi.org/project/alfie/).

Executing multi-taxa eDNA ecological assessment via traditional metrics and interactive networks

Current approaches to ecological assessment are limited by the traditional morpho-taxonomic methods presently employed and the inability to meet increasing demands for rapid assessments. Advancements in high throughput sequencing now enable rapid high-resolution ecological assessment using environmental DNA (eDNA). Here we test the ability of using eDNA-based ecological assessment methods against traditional assessment of two key indicator groups (diatoms and macroinvertebrates) and show how eDNA across multiple gene regions (COI, rbcL, 12S and 18S) can be used to infer interactive networks that link to ecological assessment criteria. We compared results between taxonomic and eDNA based assessments and found significant positive associations between macroinvertebrate (p < 0.001 R² = 0.645) and diatom (p = 0.015, R² = 0.222) assessment metrics. We further assessed the ability of eDNA based assessment to identify environmentally sensitive genera and found an order of magnitude greater potential for 18S, versus COI or rbcL, to determine environmental filtering of ecologically assessed communities. Lastly, we compared the ability of traditional metrics against co-occurrence network properties of our combined 18S, COI and rbcL indicator genera to infer habitat quality measures currently used by managers. We found that transitivity (network connectivity), linkage density and cohesion were significantly associated with habitat modification scores (HMS), whereas network properties were inconsistent with linking to the habitat quality score (HQS) metric. The incorporation of multi-marker eDNA network assessment opens up a means for finer scale ecological assessment, currently limited using traditional methods. While utilization of eDNA-based assessment is recommended, direct comparisons with traditional approaches are difficult as the methods are intrinsically different and should be treated as such with regards to future research. Overall, our findings show that eDNA can be used for effective ecological assessment while offering a wider range of scope and application compared to traditional assessment methods.

Accumulation curves of environmental DNA sequences predict coastal fish diversity in the coral triangle

Environmental DNA (eDNA) has the potential to provide more comprehensive biodiversity assessments, particularly for vertebrates in species-rich regions. However, this method requires the completeness of a reference database (i.e. a list of DNA sequences attached to each species), which is not currently achieved for many taxa and ecosystems. As an alternative, a range of operational taxonomic units (OTUs) can be extracted from eDNA metabarcoding. However, the extent to which the diversity of OTUs provided by a limited eDNA sampling effort can predict regional species diversity is unknown. Here, by modelling OTU accumulation curves of eDNA seawater samples across the Coral Triangle, we obtained an asymptote reaching 1531 fish OTUs, while 1611 fish species are recorded in the region. We also accurately predict (R² = 0.92) the distribution of species richness among fish families from OTU-based asymptotes. Thus, the multi-model framework of OTU accumulation curves extends the use of eDNA metabarcoding in ecology, biogeography and conservation.

A microbial mandala for environmental monitoring: Predicting multiple impacts on estuarine prokaryote communities of the Bay of Biscay

Routine monitoring of benthic biodiversity is critical for managing and understanding the anthropogenic impacts on marine, transitional and freshwater ecosystems. However, traditional reliance on morphological identification generally makes it cost-prohibitive to increase the scale of monitoring programmes. Metabarcoding of environmental DNA has clear potential to overcome many of the problems associated with traditional monitoring, with prokaryotes and other microorganisms showing particular promise as bioindicators. However, due to the limited knowledge regarding the ecological roles and responses of environmental microorganisms to different types of pressure, the use of de novo approaches is necessary. Here, we use two such approaches for the prediction of multiple impacts present in estuaries and coastal areas of the Bay of Biscay based on microbial communities. The first (Random Forests) is a machine learning method while the second (Threshold Indicator Taxa Analysis and quantile regression splines) is based on de novo identification of bioindicators. Our results show that both methods overlap considerably in the indicator taxa identified, but less for sequence variants. Both methods also perform well in spite of the complexity of the studied ecosystem, providing predictive models with strong correlation to reference values and fair to good agreement with ecological status groups. The ability to predict several specific types of pressure is especially appealing. The cross-validated models and biotic indices developed can be directly applied to predict the environmental status of estuaries in the same geographical region, although more work is needed to evaluate and improve them for use in new regions or habitats.

Towards Naples Ecological REsearch for Augmented Observatories (NEREA): The NEREA-Fix Module, a Stand-Alone Platform for Long-Term Deep-Sea Ecosystem Monitoring

Deep-sea ecological monitoring is increasingly recognized as indispensable for the comprehension of the largest biome on Earth, but at the same time it is subjected to growing human impacts for the exploitation of biotic and abiotic resources. Here, we present the Naples Ecological REsearch (NEREA) stand-alone observatory concept (NEREA-fix), an integrated observatory with a modular, adaptive structure, characterized by a multiparametric video-platform to be deployed in the Dohrn canyon (Gulf of Naples, Tyrrhenian Sea) at ca. 650 m depth. The observatory integrates a seabed platform with optoacoustic and oceanographic/geochemical sensors connected to a surface transmission buoy, plus a mooring line (also equipped with depth-staged environmental sensors). This reinforced high-frequency and long-lasting ecological monitoring will integrate the historical data conducted over 40 years for the Long-Term Ecological Research (LTER) at the station “Mare Chiara”, and ongoing vessel-assisted plankton (and future environmental DNA-eDNA) sampling. NEREA aims at expanding the observational capacity in a key area of the Mediterranean Sea, representing a first step towards the establishment of a bentho-pelagic network to enforce an end-to-end transdisciplinary approach for the monitoring of marine ecosystems across a wide range of animal sizes (from bacteria to megafauna).

Foresight Workshop on Advances in Ocean Biological Observations: a sustained system for deep-ocean meroplankton

Recent advances in technology have enabled an unprecedented development of underwater research, extending from near shore to the deepest regions of the globe. However, monitoring of biodiversity is not fully implemented in political agendas and biological observations in the deep ocean have been even more limited in space and time.

The Foresight Workshop on Advances in Ocean Biological Observations: a sustained system for deep-ocean meroplankton was convened to to foster advances in the knowledge on deep-ocean invertebrate larval distributions and improve our understanding of fundamental deep-ocean ecological processes such as connectivity and resilience of benthic communities to natural and human-induced disturbance. This Meroplankton Observations Workshop had two specific goals: 1) review the state-of-the-art instrumentation available for meroplankton observations; 2) develop a strategy to implement technological innovations for in-situ meroplankton observation. Presentations and discussions are summarised in this report covering: i) key challenges and priorities for advancing the knowledge of deep-sea larval diversity and distribution: ii) recent developments in technology and future needs for plankton observation, iii) data integration and oceanographic modelling; iv) synergies and added value of a sustained observation system for meroplankton; v) steps for developing a sustained observation system for deep-ocean meroplankton and plans to maximise collaborative opportunities.

Supervised machine learning is superior to indicator value inference in monitoring the environmental impacts of salmon aquaculture using eDNA metabarcodes

Increasing anthropogenic impact and global change effects on natural ecosystems has prompted the development of less expensive and more efficient bioassessments methodologies. One promising approach is the integration of DNA metabarcoding in environmental monitoring. A critical step in this process is the inference of ecological quality (EQ) status from identified molecular bioindicator signatures that mirror environmental classification based on standard macroinvertebrate surveys. The most promising approaches to infer EQ from biotic indices (BI) are supervised machine learning (SML) and the calculation of indicator values (IndVal). In this study we compared the performance of both approaches using DNA metabarcodes of bacteria and ciliates as bioindicators obtained from 152 samples collected from seven Norwegian salmon farms. Results from standard macroinvertebrate-monitoring of the same samples were used as reference to compare the accuracy of both approaches. First, SML outperformed the IndVal approach to infer EQ from eDNA metabarcodes. The Random Forest (RF) algorithm appeared to be less sensitive to noisy data (a typical feature of massive environmental sequence data sets) and uneven data coverage across EQ classes (a typical feature of environmental compliance monitoring scheme) compared to a widely used method to infer IndVals for the calculation of a BI. Second, bacteria allowed for a more accurate EQ assessment than ciliate eDNA metabarcodes. For the implementation of DNA metabarcoding into routine monitoring programmes to assess EQ around salmon aquaculture cages, we therefore recommend bacterial DNA metabarcodes in combination with SML to classify EQ categories based on molecular signatures.

Machine Learning to Evaluate Impacts of Flood Protection in Bangladesh, 1983–2014

Impacts of climate change adaptation strategies need to be evaluated using principled methods spanning sectors and longer time frames. We propose machine-learning approaches to study the long-term impacts of flood protection in Bangladesh. Available data include socio-economic survey and events data (death, migration, etc.) from 1983–2014. These multidecadal data, rare in their extent and quality, provide a basis for using machine-learning approaches even though the data were not collected or designed to assess the impact of the flood control investments. We test whether the embankment has affected the welfare of people over time, benefiting those living inside more than those living outside. Machine-learning approaches enable learning patterns in data to help discriminate between two groups: here households living inside vs. outside. They also help identify the most informative indicators of discrimination and provide robust metrics to evaluate the quality of the model. Overall, we find no significant difference between inside/outside populations based on welfare, migration, or mortality indicators. However, we note a significant difference in inward/outward movement with respect to the embankment. While certain data gaps and spatial heterogeneity in sampled populations suggest caution in any conclusive interpretation of the flood protection infrastructure, we do not see higher benefits accruing to those living with higher levels of protection. This has implications for Bangladesh’s planning for future and more extreme climate futures, including the national Delta Plan, and global investments in climate resilient infrastructure to create positive social impacts.

Machine learning for total organic carbon analysis of environmental water samples using high-throughput colorimetric sensors

Process of total organic carbon (TOC) prediction using colorimetric sensors and machine learning (ML).

eDNA metabarcoding in zooplankton improves the ecological status assessment of aquatic ecosystems

Aquatic ecosystems are monitored worldwide using a range of biological quality elements that are morphologically identified. The environmental DNA (eDNA)-based approach has unprecedented advantages (e.g., high throughput, high efficiency and low cost) for biodiversity surveys in both freshwater and marine ecosystems compared with traditional sampling and image recognition. The use of eDNA has been mostly limited to biodiversity estimation, how to apply the eDNA approach in assessing the ecological health status is largely unexplored. Here, using zooplankton as an example, we examined the application of eDNA monitoring for ecological status assessment in an aquatic ecosystem. The results showed that eDNA monitoring reflected the spatial and temporal variations in zooplankton structure. Both species composition and bio-interactions varied significantly between sampling seasons (dry, normal and wet). A total of 60 different zooplankton indices were calculated based on eDNA monitoring and most of these indices were highly correlated with the level of water pollution, which was indicated by the water quality index in one or all three seasons. Both qualitative and quantitative eDNA-based biological indices were correlated with water quality. The season-dependent eDNA zooplankton integrity index (IZI) reflected the ecological status, and this method improves the timeliness of bioassessment.

Analysis of 13,312 benthic invertebrate samples from German streams reveals minor deviations in ecological status class between abundance and presence/absence data

Benthic invertebrates are the most commonly used organisms used to assess ecological status as required by the EU Water Framework Directive (WFD). For WFD-compliant assessments, benthic invertebrate communities are sampled, identified and counted. Taxa × abundance matrices are used to calculate indices and the resulting scores are compared to reference values to determine the ecological status class. DNA-based tools, such as DNA metabarcoding, provide a new and precise method for species identification but cannot deliver robust abundance data. To evaluate the applicability of DNA-based tools to ecological status assessment, we evaluated whether the results derived from presence/absence data are comparable to those derived from abundance data. We analysed benthic invertebrate community data obtained from 13,312 WFD assessments of German streams. Broken down to 30 official stream types, we compared assessment results based on abundance and presence/absence data for the assessment modules “organic pollution” (i.e., the saprobic index) and “general degradation” (a multimetric index) as well as their underlying metrics.

In 76.6% of cases, the ecological status class did not change after transforming abundance data to presence/absence data. In 12% of cases, the status class was reduced by one (e.g., from good to moderate), and in 11.2% of cases, the class increased by one. In only 0.2% of cases, the status shifted by two classes. Systematic stream type-specific deviations were found and differences between abundance and presence/absence data were most prominent for stream types where abundance information contributed directly to one or several metrics of the general degradation module. For a single stream type, these deviations led to a systematic shift in status from ‘good’ to ‘moderate’ (n = 201; with only n = 3 increasing). The systematic decrease in scores was observed, even when considering simulated confidence intervals for abundance data. Our analysis suggests that presence/absence data can yield similar assessment results to those for abundance-based data, despite type-specific deviations. For most metrics, it should be possible to intercalibrate the two data types without substantial efforts. Thus, benthic invertebrate taxon lists generated by standardised DNA-based methods should be further considered as a complementary approach.

Biodiversity assessment of tropical shelf eukaryotic communities via pelagic eDNA metabarcoding

Our understanding of marine communities and their functions in an ecosystem relies on the ability to detect and monitor species distributions and abundances. Currently, the use of environmental DNA (eDNA) metabarcoding is increasingly being applied for the rapid assessment and monitoring of aquatic species. Most eDNA metabarcoding studies have either focussed on the simultaneous identification of a few specific taxa/groups or have been limited in geographical scope. Here, we employed eDNA metabarcoding to compare beta diversity patterns of complex pelagic marine communities in tropical coastal shelf habitats spanning the whole Caribbean Sea. We screened 68 water samples using a universal eukaryotic COI barcode region and detected highly diverse communities, which varied significantly among locations, and proved good descriptors of habitat type and environmental conditions. Less than 15% of eukaryotic taxa were assigned to metazoans, most DNA sequences belonged to a variety of planktonic "protists," with over 50% of taxa unassigned at the phylum level, suggesting that the sampled communities host an astonishing amount of micro-eukaryotic diversity yet undescribed or absent from COI reference databases. Although such a predominance of micro-eukaryotes severely reduces the efficiency of universal COI markers to investigate vertebrate and other metazoans from aqueous eDNA, the study contributes to the advancement of rapid biomonitoring methods and brings us closer to a full inventory of extant marine biodiversity.

New High-Tech Flexible Networks for the Monitoring of Deep-Sea Ecosystems

Increasing interest in the acquisition of biotic and abiotic resources from within the deep sea (e.g., fisheries, oil-gas extraction, and mining) urgently imposes the development of novel monitoring technologies, beyond the traditional vessel-assisted, time-consuming, high-cost sampling surveys. The implementation of permanent networks of seabed and water-column-cabled (fixed) and docked mobile platforms is presently enforced, to cooperatively measure biological features and environmental (physicochemical) parameters. Video and acoustic (i.e., optoacoustic) imaging are becoming central approaches for studying benthic fauna (e.g., quantifying species presence, behavior, and trophic interactions) in a remote, continuous, and prolonged fashion. Imaging is also being complemented by in situ environmental-DNA sequencing technologies, allowing the traceability of a wide range of organisms (including prokaryotes) beyond the reach of optoacoustic tools. Here, we describe the different fixed and mobile platforms of those benthic and pelagic monitoring networks, proposing at the same time an innovative roadmap for the automated computing of hierarchical ecological information on deep-sea ecosystems (i.e., from single species' abundance and life traits to community composition, and overall biodiversity).

Application of Machine Learning in Microbiology

Microorganisms are ubiquitous and closely related to people's daily lives. Since they were first discovered in the 19th century, researchers have shown great interest in microorganisms. People studied microorganisms through cultivation, but this method is expensive and time consuming. However, the cultivation method cannot keep a pace with the development of high-throughput sequencing technology. To deal with this problem, machine learning (ML) methods have been widely applied to the field of microbiology. Literature reviews have shown that ML can be used in many aspects of microbiology research, especially classification problems, and for exploring the interaction between microorganisms and the surrounding environment. In this study, we summarize the application of ML in microbiology.

Multi-marker eDNA metabarcoding survey to assess the environmental impact of three offshore gas platforms in the North Adriatic Sea (Italy)

The environmental DNA (eDNA) metabarcoding represents a new promising tool for biomonitoring and environmental impact assessment. One of the main advantages of eDNA metabarcoding, compared to the traditional morphotaxonomy-based methods, is to provide a more holistic biodiversity information that includes inconspicuous morphologically non-identifiable taxa. Here, we use eDNA metabarcoding to survey marine biodiversity in the vicinity of the three offshore gas platforms in North Adriatic Sea (Italy). We isolated eDNA from 576 water and sediment samples collected at 32 sampling sites situated along four axes at increasing distances from the gas platforms. We obtained about 46 million eDNA sequences for 5 markers from nuclear 18S V1V2, 18S V4, 18S 37F and mitochondrial 16S and COI genes that cover a wide diversity of benthic and planktonic eukaryotes. Our results showed some impact of platform activities on benthic and pelagic communities at very close distance (<50 m), while communities for intermediate (125 m, 250 m, 500 m) and reference (1000 m, 2000 m) sites did not show any particular biodiversity changes that could be related to platforms activities. The most significant community change along the distance gradient was obtained with the 18S V1V2 marker targeting benthic eukaryotes, even though other markers showed similar trends, but to a lesser extent. These results were congruent with the AMBI index inferred from the eDNA sequences assigned to benthic macrofauna. We finally explored the relation between various physicochemical parameters, including hydrocarbons, on benthic community in the case of one of the platforms. Our results showed that these communities were not significantly impacted by most of hydrocarbons, but rather by macro-elements and sediment texture.

Using Bayesian networks to predict risk to estuary water quality and patterns of benthic environmental DNA in Queensland

Predictive modeling can inform natural resource management by representing stressor-response pathways in a logical way and quantifying the effects on selected endpoints. This study demonstrates a risk assessment model using the Bayesian network relative risk model (BN-RRM) approach to predict water quality and, for the first time, eukaryote environmental DNA (eDNA) data as a measure of benthic community structure. Environmental DNA sampling is a technique for biodiversity measurements that involves extracting DNA from environmental samples, amplicon sequencing a targeted gene, in this case the 18s rDNA gene (which targets eukaryotes), and matching the sequences to organisms. Using a network of probability distributions, the BN-RRM model predicts risk to water quality objectives and the relative richness of benthic taxa groups in the Noosa, Pine, and Logan estuaries in Southeast Queensland (SEQ), Australia. The model predicts Dissloved Oxygen more accurately than the chlorophyll a water quality endpoint and photosynthesizing benthos more accurately than heterotrophs. Results of BN-RRM modeling given current inputs indicate that the water quality and benthic assemblages of the Noosa are relatively homogenous across all sub risk regions, and that the Noosa has a 73%-92% probability of achieving water quality objectives, indicating a low relative risk. Conversely, the Middle Logan, Middle Pine, and Lower Pine regions are much less likely to meet objectives (15%-55% probability), indicating a relatively higher risk to water quality in those regions. The benthic community richness patterns associated with risk in the Noosa are high Diatom relative richness and low Green Algae relative richness. The only benthic pattern consistently associated with the relatively higher risk to water quality is high richness of fungi species. The BN-RRM model provides a basis for future predictions and adaptive management at the direction of resource managers. Integr Environ Assess Manag 2019;15:93-111. © 2018 SETAC.

Development and implementation of eco-genomic tools for aquatic ecosystem biomonitoring: the SYNAQUA French-Swiss program

The effectiveness of environmental protection measures is based on the early identification and diagnosis of anthropogenic pressures. Similarly, restoration actions require precise monitoring of changes in the ecological quality of ecosystems, in order to highlight their effectiveness. Monitoring the ecological quality relies on bioindicators, which are organisms revealing the pressures exerted on the environment through the composition of their communities. Their implementation, based on the morphological identification of species, is expensive because it requires time and experts in taxonomy. Recent genomic tools should provide access to reliable and high-throughput environmental monitoring by directly inferring the composition of bioindicators' communities from their DNA (metabarcoding). The French-Swiss program SYNAQUA (INTERREG France-Switzerland 2017-2019) proposes to use and validate the tools of environmental genomic for biomonitoring and aims ultimately at their implementation in the regulatory bio-surveillance. SYNAQUA will test the metabarcoding approach focusing on two bioindicators, diatoms, and aquatic oligochaetes, which are used in freshwater biomonitoring in France and Switzerland. To go towards the renewal of current biomonitoring practices, SYNAQUA will (1) bring together different actors: scientists, environmental managers, consulting firms, and biotechnological companies, (2) apply this approach on a large scale to demonstrate its relevance, (3) propose robust and reliable tools, and (4) raise public awareness and train the various actors likely to use these new tools. Biomonitoring approaches based on such environmental genomic tools should address the European need for reliable, higher-throughput monitoring to improve the protection of aquatic environments under multiple pressures, guide their restoration, and follow their evolution.

Uses and Misuses of Environmental DNA in Biodiversity Science and Conservation

The study of environmental DNA (eDNA) has the potential to revolutionize biodiversity science and conservation action by enabling the census of species on a global scale in near real time. To achieve this promise, technical challenges must be resolved. In this review, we explore the main uses of eDNA as well as the complexities introduced by its misuse. Current eDNA methods require refinement and improved calibration and validation along the entire workflow to lessen false positives/negatives. Moreover, there is great need for a better understanding of the “natural history” of eDNA—its origins, state, lifetime, and transportation—and for more detailed insights concerning the physical and ecological limitations of eDNA use. Although eDNA analysis can provide powerful information, particularly in freshwater and marine environments, its impact is likely to be less significant in terrestrial settings. The broad adoption of eDNA tools in conservation will largely depend on addressing current uncertainties in data interpretation.

Towards an eDNA metabarcode-based performance indicator for full-scale municipal wastewater treatment plants

Knowledge-driven management for wastewater treatment plant (WWTP) performance becomes increasingly important considering the globally growing production of wastewater and the rising demand of clean water supply. Even though the potential of microbial organisms (bacteria and protists) as bioindicators for WWTP performance is well known, it is far from being fully exploited for routine monitoring programs. Therefore, we here used massive sequencing of environmental (e)DNA metabarcodes from bacterial (V3-V4 region of the SSU rRNA gene) and eukaryote (V9 region of the SSU rRNA gene) communities in 21 activated sludge samples obtained from full-scale municipal WWTPs in Germany. Microbial community patterns were compared to standard WWTP operating parameters and two traditionally used WWTP performance indicators (Sludge Biotic Index and Sludge Index). Both indices showed low concordance and hardly correlated with chemical WWTP performance parameters nor did they correlate with microbial community structures. In contrast, microbial community profiles significantly correlated with WWTP performance parameters and operating conditions of the plants under study. Therefore, eDNA metabarcode profiles of whole microbial communities indicate the performance of WWTP and can provide useful information for management strategies. We here suggest a strategy for the development of an eDNA metabarcode based bioindicator system, which can be implemented in future standard monitoring programs for WWTP performance and effluent quality.

Application of Environmental DNA Metabarcoding for Predicting Anthropogenic Pollution in Rivers

Rivers are among the most threatened freshwater ecosystems, and anthropogenic activities are affecting both river structures and water quality. While assessing the organisms can provide a comprehensive measure of a river's ecological status, it is limited by the traditional morphotaxonomy-based biomonitoring. Recent advances in environmental DNA (eDNA) metabarcoding allow to identify prokaryotes and eukaryotes in one sequencing run, and could thus allow unprecedented resolution. Whether such eDNA-based data can be used directly to predict the pollution status of rivers as a complementation of environmental data remains unknown. Here we used eDNA metabarcoding to explore the main stressors of rivers along which community structure changes, and to identify the method's potential for predicting pollution status based on eDNA data. We showed that a broad range of taxa in bacterial, protistan, and metazoan communities could be profiled with eDNA. Nutrients were the main driving stressor affecting communities' structure, alpha diversity, and the ecological network. We specifically observed that the relative abundance of indicative OTUs was significantly correlated with nutrient levels. These OTUs data could be used to predict the nutrient status up to 79% accuracy on testing data sets. Thus, our study gives a novel approach to predicting the pollution status of rivers by eDNA data.

The future of biotic indices in the ecogenomic era: Integrating (e)DNA metabarcoding in biological assessment of aquatic ecosystems

The bioassessment of aquatic ecosystems is currently based on various biotic indices that use the occurrence and/or abundance of selected taxonomic groups to define ecological status. These conventional indices have some limitations, often related to difficulties in morphological identification of bioindicator taxa. Recent development of DNA barcoding and metabarcoding could potentially alleviate some of these limitations, by using DNA sequences instead of morphology to identify organisms and to characterize a given ecosystem. In this paper, we review the structure of conventional biotic indices, and we present the results of pilot metabarcoding studies using environmental DNA to infer biotic indices. We discuss the main advantages and pitfalls of metabarcoding approaches to assess parameters such as richness, abundance, taxonomic composition and species ecological values, to be used for calculation of biotic indices. We present some future developments to fully exploit the potential of metabarcoding data and improve the accuracy and precision of their analysis. We also propose some recommendations for the future integration of DNA metabarcoding to routine biomonitoring programs.

Comparison of environmental DNA and bulk-sample metabarcoding using highly degenerate cytochrome c oxidase I primers

Freshwater biodiversity provides important ecosystem services and is at the core of water quality monitoring worldwide. To assess freshwater biodiversity, genetic methods such as metabarcoding are increasingly used as they are faster and allow better taxonomic resolution than manual identification methods. Either sampled organisms are used directly for "bulk metabarcoding," or water is filtered and the extracted environmental DNA serves as a proxy for biodiversity via "eDNA metabarcoding." Despite the advantages of both methods, questions remain regarding their comparability and applicability for routine biomonitoring and stressor impact assessment. Therefore, we compared metabarcoding results from bulk and eDNA samples taken from 19 streams spanning a wide gradient of farming intensities in New Zealand. We performed PCR with highly degenerate cytochrome c oxidase I primers and sequenced libraries on an Illumina MiSeq. The inferred community composition differed strongly between the two methods. More taxa were captured by eDNA than bulk-sample metabarcoding (5,819 vs. 1,483), but more of the commonly used invertebrate bioindicator taxa (mayflies, stoneflies and caddisflies) were found in bulk (47) than eDNA samples (37). Catchment-wide and local land use impacts on communities were detected better by eDNA metabarcoding, especially for non-metazoan taxa. Our findings imply that bulk-sample metabarcoding resembles classical freshwater biomonitoring approaches better, as more indicator macroinvertebrate taxa are captured. However, eDNA metabarcoding might be better suited to infer the impact of stressors on stream ecosystems at larger scales, as many new and potentially more informative taxa are registered. We therefore suggest exploring both methods in future assessments of stream biodiversity.

Ballast Water Exchange and Invasion Risk Posed by Intracoastal Vessel Traffic: An Evaluation Using High Throughput Sequencing

Ballast water remains a potent vector of non-native aquatic species introductions, despite increased global efforts to reduce risk of ballast water mediated invasions. This is particularly true of intracoastal vessel traffic, whose characteristics may limit the feasibility and efficacy of management through ballast water exchange (BWE). Here we utilize high throughput sequencing (HTS) to assess biological communities associated with ballast water being delivered to Valdez, Alaska from multiple source ports along the Pacific Coast of the United States. Our analyses indicate that BWE has a significant but modest effect on ballast water assemblages. Although overall richness was not reduced with exchange, we detected losses of some common benthic coastal taxa (e.g., decapods, mollusks, bryozoans, cnidaria) and gains of open ocean taxa (e.g., certain copepods, diatoms, and dinoflagellates), including some potentially toxic species. HTS-based metabarcoding identified significantly differentiated biodiversity signatures from individual source ports; this signal persisted, though weakened, in vessels undergoing BWE, indicating incomplete faunal turnover associated with management. Our analysis also enabled identification of taxa that may be of particular concern if established in Alaskan waters. While these results reveal a clear effect of BWE on diversity in intracoastal transit, they also indicate continued introduction risk of non-native and harmful taxa.

A DNA metabarcoding protocol for hyporheic freshwater meiofauna: Evaluating highly degenerate COI primers and replication strategy

The hyporheic zone, i.e. the ecotone between surface water and the groundwater, is a rarely studied freshwater ecosystem. Hyporheic taxa are often meiofaunal (&lt;1 mm) in size and difficult to identify based on morphology. Metabarcoding approaches are promising for the study of these environments and taxa, but it is yet unclear if commonly applied metabarcoding primers and replication strategies can be used. In this study, we took sediment cores from two near natural upstream (NNU) and two ecologically improved downstream (EID) sites in the Boye catchment (Emscher River, Germany), metabarcoding their meiofaunal communities. We evaluated the usability of a commonly used, highly degenerate COI primer pair (BF2/BR2) and tested how sequencing three PCR replicates per sample and removing MOTUs present in only one out of three replicates impacts the inferred community composition. A total of 22,514 MOTUs were detected, of which only 263 were identified as Metazoa. Our results highlight the gaps in reference databases for meiofaunal taxa and the potential problems of using highly degenerate primers for studying samples containing a high number of non-metazoan taxa. Alpha diversity was higher in EID sites and showed higher community similarity when compared to NNU sites. Beta diversity analyses showed that removing MOTUs detected in only one out of three replicates per site greatly increased community similarity in samples. Sequencing three sample replicates and removing rare MOTUs is seen as a good compromise between retaining too many false-positives and introducing too many false-negatives. We conclude that metabarcoding hyporheic communities using highly degenerate COI primers can provide valuable first insights into the diversity of these ecosystems and highlight some potential application scenarios.

Boosting DNA metabarcoding for biomonitoring with phylogenetic estimation of operational taxonomic units' ecological profiles

DNA metabarcoding has been introduced as a revolutionary way to identify organisms and monitor ecosystems. However, the potential of this approach for biomonitoring remains partially unfulfilled because a significant part of the sampled DNA cannot be affiliated to species due to incomplete reference libraries. Thus, biotic indices, which are based on the estimated abundances of species in a community and their ecological profiles, can be inaccurate. We propose to compute biotic indices using phylogenetic imputation of operational taxonomic units (OTUs') ecological profiles (OTU-PITI approach). First, OTUs sequences are inserted within a reference phylogeny. Second, OTUs' ecological profiles are estimated on the basis of their phylogenetic relationships with reference species whose ecology is known. Based on these ecological profiles, biotic indices can be computed using all available OTUs. Using freshwater diatoms as a case study, we show that short DNA barcodes can be placed accurately within a phylogeny and their ecological preferences estimated with a satisfactory level of precision. In the light of these results, we tested the approach with a data set of 139 environmental samples of benthic river diatoms for which the same biotic index (specific sensitivity index) was calculated using (a) traditional microscopy, (b) OTUs with taxonomic assignment approach, (c) OTUs with phylogenetic estimation of ecological profiles (OTU-PITI) and (d) OTU with taxonomic assignment completed by the phylogenetic approach (OTU-PITI) for unclassified OTUs. Using traditional microscopy as a reference, we found that the combination of the OTUs' taxonomic assignment completed by the phylogenetic method performed satisfactorily and substantially better than the other methods tested.

BBI: an R package for the computation of Benthic Biotic Indices from composition data

The monitoring of impacts of anthropic activities in marine environments, such as aquaculture, oil-drilling platforms or deep-sea mining, relies on Benthic Biotic Indices (BBI). Several indices have been formalised to reduce the multivariate composition data into a single continuous value that is ascribed to a discrete ecological quality status. Such composition data is traditionally obtained from macrofaunal inventories, which is time-consuming and expertise-demanding. Important efforts are ongoing towards using High-Throughput Sequencing of environmental DNA (eDNA metabarcoding) to replace or complement morpho-taxonomic surveys for routine biomonitoring. The computation of BBI from such composition data is usually being undertaken by practitioners with excel spreadsheets or through custom script. Hence, the updating of reference morpho-taxonomic tables and cross studies comparison could be hampered. Here we introduce the R package BBI for the computation of BBI from composition data, either obtained from traditional morpho-taxonomic inventories or from metabarcoding data. Its aim is to provide an open-source, transparent and centralised method to compute BBI for routine biomonitoring.

MicroPheno: predicting environments and host phenotypes from 16S rRNA gene sequencing using a k-mer based representation of shallow sub-samples

Motivation

Microbial communities play important roles in the function and maintenance of various biosystems, ranging from the human body to the environment. A major challenge in microbiome research is the classification of microbial communities of different environments or host phenotypes. The most common and cost-effective approach for such studies to date is 16S rRNA gene sequencing. Recent falls in sequencing costs have increased the demand for simple, efficient and accurate methods for rapid detection or diagnosis with proved applications in medicine, agriculture and forensic science. We describe a reference- and alignment-free approach for predicting environments and host phenotypes from 16S rRNA gene sequencing based on k-mer representations that benefits from a bootstrapping framework for investigating the sufficiency of shallow sub-samples. Deep learning methods as well as classical approaches were explored for predicting environments and host phenotypes.

Results

A k-mer distribution of shallow sub-samples outperformed Operational Taxonomic Unit (OTU) features in the tasks of body-site identification and Crohn's disease prediction. Aside from being more accurate, using k-mer features in shallow sub-samples allows (i) skipping computationally costly sequence alignments required in OTU-picking and (ii) provided a proof of concept for the sufficiency of shallow and short-length 16S rRNA sequencing for phenotype prediction. In addition, k-mer features predicted representative 16S rRNA gene sequences of 18 ecological environments, and 5 organismal environments with high macro-F1 scores of 0.88 and 0.87. For large datasets, deep learning outperformed classical methods such as Random Forest and Support Vector Machine.

Availability and implementation

The software and datasets are available at https://llp.berkeley.edu/micropheno.

Supplementary information

Supplementary data are available at Bioinformatics online.

DNA metabarcoding of littoral hard-bottom communities: high diversity and database gaps revealed by two molecular markers

Biodiversity assessment of marine hard-bottom communities is hindered by the high diversity and size-ranges of the organisms present. We developed a DNA metabarcoding protocol for biodiversity characterization of structurally complex natural marine hard-bottom communities. We used two molecular markers: the “Leray fragment” of mitochondrial cytochrome c oxidase (COI), for which a novel primer set was developed, and the V7 region of the nuclear small subunit ribosomal RNA (18S). Eight different shallow marine littoral communities from two National Parks in Spain (one in the Atlantic Ocean and another in the Mediterranean Sea) were studied. Samples were sieved into three size fractions from where DNA was extracted separately. Bayesian clustering was used for delimiting molecular operational taxonomic units (MOTUs) and custom reference databases were constructed for taxonomic assignment. Despite applying stringent filters, we found high values for MOTU richness (2,510 and 9,679 MOTUs with 18S and COI, respectively), suggesting that these communities host a large amount of yet undescribed eukaryotic biodiversity. Significant gaps are still found in sequence reference databases, which currently prevent the complete taxonomic assignment of the detected sequences. In our dataset, 85% of 18S MOTUs and 64% of COI MOTUs could be identified to phylum or lower taxonomic level. Nevertheless, those unassigned were mostly rare MOTUs with low numbers of reads, and assigned MOTUs comprised over 90% of the total sequence reads. The identification rate might be significantly improved in the future, as reference databases are further completed. Our results show that marine metabarcoding, currently applied mostly to plankton or sediments, can be adapted to structurally complex hard bottom samples. Thus, eukaryotic metabarcoding emerges as a robust, fast, objective and affordable method to comprehensively characterize the diversity of marine benthic communities dominated by macroscopic seaweeds and colonial or modular sessile metazoans. The 18S marker lacks species-level resolution and thus cannot be recommended to assess the detailed taxonomic composition of these communities. Our new universal primers for COI can potentially be used for biodiversity assessment with high taxonomic resolution in a wide array of marine, terrestrial or freshwater eukaryotic communities.

Omics Advances in Ecotoxicology

Toxic substances in the environment generate adverse effects at all levels of biological organization from the molecular level to community and ecosystem. Given this complexity, it is not surprising that ecotoxicologists have struggled to address the full consequences of toxic substance release at ecosystem level, due to the limits of observational and experimental tools to reveal the changes in deep structure at different levels of organization. -Omics technologies, consisting of genomics and ecogenomics, have the power to reveal, in unprecedented detail, the cellular processes of an individual or biodiversity of a community in response to environmental change with high sample/observation throughput. This represents a historic opportunity to transform the way we study toxic substances in ecosystems, through direct linkage of ecological effects with the systems biology of organisms. Three recent examples of -omics advance in the assessment of toxic substances are explored here: (1) the use of functional genomics in the discovery of novel molecular mechanisms of toxicity of chemicals in the environment; (2) the development of laboratory pipelines of dose-dependent, reduced transcriptomics to support high-throughput chemical testing at the biological pathway level; and (3) the use of eDNA metabarcoding approaches for assessing chemical effects on biological communities in mesocosm experiments and through direct observation in field monitoring. -Omics advances in ecotoxicological studies not only generate new knowledge regarding mechanisms of toxicity and environmental effect, improving the relevance and immediacy of laboratory toxicological assessment, but can provide a wholly new paradigm for ecotoxicology by linking ecological models to mechanism-based, systems biology approaches.

Assessing the effect of mercury pollution on cultured benthic foraminifera community using morphological and eDNA metabarcoding approaches

Mercury (Hg) is a highly toxic element for living organisms and is known to bioaccumulate and biomagnify. Here, we analyze the response of benthic foraminifera communities cultured in mesocosm and exposed to different concentrations of Hg. Standard morphological analyses and environmental DNA metabarcoding show evidence that Hg pollution has detrimental effects on benthic foraminifera. The molecular analysis provides a more complete view of foraminiferal communities including the soft-walled single-chambered monothalamiids and small-sized hard-shelled rotaliids and textulariids than the morphological one. Among these taxa that are typically overlooked in morphological studies we found potential bioindicators of Hg pollution. The mesocosm approach proves to be an effective method to study benthic foraminiferal responses to various types and concentrations of pollutants over time. This study further supports foraminiferal metabarcoding as a complementary and/or alternative method to standard biomonitoring program based on the morphological identification of species communities.

Marine microbes in 4D-using time series observation to assess the dynamics of the ocean microbiome and its links to ocean health

Microbial observation is of high relevance in assessing marine phenomena of scientific and societal concern including ocean productivity, harmful algal blooms, and pathogen exposure. However, we have yet to realise its potential to coherently and comprehensively report on global ocean status. The ability of satellites to monitor the distribution of phytoplankton has transformed our appreciation of microbes as the foundation of key ecosystem services; however, more in-depth understanding of microbial dynamics is needed to fully assess natural and anthropogenically induced variation in ocean ecosystems. While this first synthesis shows that notable efforts exist, vast regions such as the ocean depths, the open ocean, the polar oceans, and most of the Southern Hemisphere lack consistent observation. To secure a coordinated future for a global microbial observing system, existing long-term efforts must be better networked to generate shared bioindicators of the Global Ocean's state and health.

Environmental DNA metabarcoding of benthic bacterial communities indicates the benthic footprint of salmon aquaculture

We evaluated benthic bacterial communities as bioindicators in environmental impact assessments of salmon aquaculture, a rapidly growing sector of seafood industry. Sediment samples (n=72) were collected from below salmon cages towards distant reference sites. Bacterial community profiles inferred from DNA metabarcodes were compared to reference data from standard macrofauna biomonitoring surveys of the same samples. Deltaproteobacteria were predominant in immediate vicinity of the salmon cages. Along the transect, significant shifts in bacterial community structures were observed with Gammaproteobacteria dominating the less-impacted sites. Alpha- and beta-diversity measures of bacterial communities correlated significantly with macrofauna diversity metrics and with five ecological status indices. Benthic bacterial communities mirror the reaction of macrofauna bioindicators to environmental disturbances caused by salmon farming. The implementation of bacterial eDNA metabarcoding in future Strategic Framework Directives is an alternative cost-effective high-throughput biomonitoring solution, providing a basis for management strategies in a matter of days rather than months.

Towards a holistic and solution-oriented monitoring of chemical status of European water bodies: how to support the EU strategy for a non-toxic environment?

The definition of priority substances (PS) according to the Water Framework Directive (WFD) helped to remove many of these chemicals from the market and to reduce their concentrations in the European water bodies. However, it could not prevent that many of these chemicals have been replaced by others with similar risks. Today, monitoring of the PS-based chemical status according to WFD covers only a tiny fraction of toxic risks, extensively ignores mixture effects and lacks incentives and guidance for abatement. Thus, we suggest complement this purely status-related approach with more holistic and solution-oriented monitoring, which at the same time helps to provide links to the ecological status. Major elements include (1) advanced chemical screening techniques supporting mixture risk assessment and unraveling of source-related patterns in complex mixtures, (2) effect-based monitoring for the detection of groups of chemicals with similar effects and the establishment of toxicity fingerprints, (3) effect-directed analysis of drivers of toxicity and (4) to translate chemical and toxicological fingerprints into chemical footprints for prioritization of management measures. The requirement of more holistic and solution-oriented monitoring of chemical contamination is supported by the significant advancement of appropriate monitoring tools within the last years. Non-target screening technology, effect-based monitoring and basic understanding of mixture assessment are available conceptually and in research but also increasingly find their way into practical monitoring. Substantial progress in the development, evaluation and demonstration of these tools, for example, in the SOLUTIONS project enhanced their acceptability. Further advancement, integration and demonstration, extensive data exchange and closure of remaining knowledge gaps are suggested as high priority research needs for the next future to bridge the gap between insufficient ecological status and cost-efficient abatement measures.