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

Monitoring biodiversity is essential to assess the impacts of increasing anthropogenic activities in marine environments. Traditionally, marine biomonitoring involves the sorting and morphological identification of benthic macro-invertebrates, which is time-consuming and taxonomic-expertise demanding. High-throughput amplicon sequencing of environmental DNA (eDNA metabarcoding) represents a promising alternative for benthic monitoring. However, an important fraction of eDNA sequences remains unassigned or belong to taxa of unknown ecology, which prevent their use for assessing the ecological quality status. Here, we show that supervised machine learning (SML) can be used to build robust predictive models for benthic monitoring, regardless of the taxonomic assignment of eDNA sequences. We tested three SML approaches to assess the environmental impact of marine aquaculture using benthic foraminifera eDNA, a group of unicellular eukaryotes known to be good bioindicators, as features to infer macro-invertebrates based biotic indices. We found similar ecological status as obtained from macro-invertebrates inventories. We argue that SML approaches could overcome and even bypass the cost and time-demanding morpho-taxonomic approaches in future biomonitoring.

Accurate multiplexing and filtering for high-throughput amplicon-sequencing.

Tagging amplicons with tag sequences appended to PCR primers allow the multiplexing of numerous samples for high-throughput sequencing (HTS). This approach is routinely used in HTS-based diversity analyses, especially in microbial ecology and biomedical diagnostics. However, amplicon library preparation is subject to pervasive sample sequence cross-contaminations as a result of tag switching events referred to as mistagging. Here, we sequenced seven amplicon libraries prepared using various multiplexing designs in order to measure the magnitude of this phenomenon and its impact on diversity analyses. Up to 28.2% of the unique sequences correspond to undetectable (critical) mistags in single- or saturated double-tagging libraries. We show the advantage of multiplexing samples following Latin Square Designs in order to optimize the detection of mistags and maximize the information on their distribution across samples. We use this information in designs incorporating PCR replicates to filter the critical mistags and to recover the exact composition of mock community samples. Being parameter-free and data-driven, our approach can provide more accurate and reproducible HTS data sets, improving the reliability of their interpretations.

Environmental Monitoring: Inferring the Diatom Index from Next-Generation Sequencing Data.

Diatoms are widely used as bioindicators for the assessment of water quality in rivers and streams. Classically, the diatom biotic indices are based on the relative abundance of morphologically identified species weighted by their autoecological value. Obtaining such indices is time-consuming, costly, and requires excellent taxonomic expertise, which is not always available. Here we tested the possibility to overcome these limitations using a next-generation sequencing (NGS) approach to identify and quantify diatoms found in environmental DNA and RNA samples. We analyzed 27 river sites in the Geneva area (Switzerland), in order to compare the values of the Swiss Diatom Index (DI-CH) computed either by microscopic quantification of diatom species or directly from NGS data. Despite gaps in the reference database and variations in relative abundance of analyzed species, the diatom index shows a significant correlation between morphological and molecular data indicating similar biological quality status for the majority of sites. This proof-of-concept study demonstrates the potential of the NGS approach for identification and quantification of diatoms in environmental samples, opening new avenues toward the routine application of genetic tools for bioassessment and biomonitoring of aquatic ecosystems.

Ultra-deep sequencing of foraminiferal microbarcodes unveils hidden richness of early monothalamous lineages in deep-sea sediments.

Deep-sea floors represent one of the largest and most complex ecosystems on Earth but remain essentially unexplored. The vastness and remoteness of this ecosystem make deep-sea sampling difficult, hampering traditional taxonomic observations and diversity assessment. This problem is particularly true in the case of the deep-sea meiofauna, which largely comprises small-sized, fragile, and difficult-to-identify metazoans and protists. Here, we introduce an ultra-deep sequencing-based metagenetic approach to examine the richness of benthic foraminifera, a principal component of deep-sea meiofauna. We used Illumina sequencing technology to assess foraminiferal richness in 31 unsieved deep-sea sediment samples from five distinct oceanic regions. We sequenced an extremely short fragment (36 bases) of the small subunit ribosomal DNA hypervariable region 37f, which has been shown to accurately distinguish foraminiferal species. In total, we obtained 495,978 unique sequences that were grouped into 1,643 operational taxonomic units, of which about half (841) could be reliably assigned to foraminifera. The vast majority of the operational taxonomic units (nearly 90%) were either assigned to early (ancient) lineages of soft-walled, single-chambered (monothalamous) foraminifera or remained undetermined and yet possibly belong to unknown early lineages. Contrasting with the classical view of multichambered taxa dominating foraminiferal assemblages, our work reflects an unexpected diversity of monothalamous lineages that are as yet unknown using conventional micropaleontological observations. Although we can only speculate about their morphology, the immense richness of deep-sea phylotypes revealed by this study suggests that ultra-deep sequencing can improve understanding of deep-sea benthic diversity considered until now as unknowable based on a traditional taxonomic approach.

Ancient DNA complements microfossil record in deep-sea subsurface sediments.

Deep-sea subsurface sediments are the most important archives of marine biodiversity. Until now, these archives were studied mainly using the microfossil record, disregarding large amounts of DNA accumulated on the deep-sea floor. Accessing ancient DNA (aDNA) molecules preserved down-core would offer unique insights into the history of marine biodiversity, including both fossilized and non-fossilized taxa. Here, we recover aDNA of eukaryotic origin across four cores collected at abyssal depths in the South Atlantic, in up to 32.5 thousand-year-old sediment layers. Our study focuses on Foraminifera and Radiolaria, two major groups of marine microfossils also comprising diverse non-fossilized taxa. We describe their assemblages in down-core sediment layers applying both micropalaeontological and environmental DNA sequencing approaches. Short fragments of the foraminiferal and radiolarian small subunit rRNA gene recovered from sedimentary DNA extracts provide evidence that eukaryotic aDNA is preserved in deep-sea sediments encompassing the last glacial maximum. Most aDNA were assigned to non-fossilized taxa that also dominate in molecular studies of modern environments. Our study reveals the potential of aDNA to better document the evolution of past marine ecosystems and opens new horizons for the development of deep-sea palaeogenomics.

Taxonomy-free molecular diatom index for high-throughput eDNA biomonitoring.

Current biodiversity assessment and biomonitoring are largely based on the morphological identification of selected bioindicator taxa. Recently, several attempts have been made to use eDNA metabarcoding as an alternative tool. However, until now, most applied metabarcoding studies have been based on the taxonomic assignment of sequences that provides reference to morphospecies ecology. Usually, only a small portion of metabarcoding data can be used due to a limited reference database and a lack of phylogenetic resolution. Here, we investigate the possibility to overcome these limitations using a taxonomy-free approach that allows the computing of a molecular index directly from eDNA data without any reference to morphotaxonomy. As a case study, we use the benthic diatoms index, commonly used for monitoring the biological quality of rivers and streams. We analysed 87 epilithic samples from Swiss rivers, the ecological status of which was established based on the microscopic identification of diatom species. We compared the diatom index derived from eDNA data obtained with or without taxonomic assignment. Our taxonomy-free approach yields promising results by providing a correct assessment for 77% of examined sites. The main advantage of this method is that almost 95% of OTUs could be used for index calculation, compared to 35% in the case of the taxonomic assignment approach. Its main limitations are under-sampling and the need to calibrate the index based on the microscopic assessment of diatoms communities. However, once calibrated, the taxonomy-free molecular index can be easily standardized and applied in routine biomonitoring, as a complementary tool allowing fast and cost-effective assessment of the biological quality of watercourses.

High-throughput sequencing and morphology perform equally well for benthic monitoring of marine ecosystems.

Environmental diversity surveys are crucial for the bioassessment of anthropogenic impacts on marine ecosystems. Traditional benthic monitoring relying on morphotaxonomic inventories of macrofaunal communities is expensive, time-consuming and expertise-demanding. High-throughput sequencing of environmental DNA barcodes (metabarcoding) offers an alternative to describe biological communities. However, whether the metabarcoding approach meets the quality standards of benthic monitoring remains to be tested. Here, we compared morphological and eDNA/RNA-based inventories of metazoans from samples collected at 10 stations around a fish farm in Scotland, including near-cage and distant zones. For each of 5 replicate samples per station, we sequenced the V4 region of the 18S rRNA gene using the Illumina technology. After filtering, we obtained 841,766 metazoan sequences clustered in 163 Operational Taxonomic Units (OTUs). We assigned the OTUs by combining local BLAST searches with phylogenetic analyses. We calculated two commonly used indices: the Infaunal Trophic Index and the AZTI Marine Biotic Index. We found that the molecular data faithfully reflect the morphology-based indices and provides an equivalent assessment of the impact associated with fish farms activities. We advocate that future benthic monitoring should integrate metabarcoding as a rapid and accurate tool for the evaluation of the quality of marine benthic ecosystems.

Environmental monitoring through protist next-generation sequencing metabarcoding: assessing the impact of fish farming on benthic foraminifera communities.

The measurement of species diversity represents a powerful tool for assessing the impacts of human activities on marine ecosystems. Traditionally, the impact of fish farming on the coastal environment is evaluated by monitoring the dynamics of macrobenthic infaunal populations. However, taxonomic sorting and morphology-based identification of the macrobenthos demand highly trained specialists and are extremely time-consuming and costly, making it unsuitable for large-scale biomonitoring efforts involving numerous samples. Here, we propose to alleviate this laborious task by developing protist metabarcoding tools based on next-generation sequencing (NGS) of environmental DNA and RNA extracted from sediment samples. In this study, we analysed the response of benthic foraminiferal communities to the variation of environmental gradients associated with salmon farms in Scotland. We investigated the foraminiferal diversity based on ribosomal minibarcode sequences generated by the Illumina NGS technology. We compared the molecular data with morphospecies counts and with environmental gradients, including distance to cages and redox used as a proxy for sediment oxygenation. Our study revealed high variations between foraminiferal communities collected in the vicinity of fish farms and at distant locations. We found evidence for species richness decrease in impacted sites, especially visible in the RNA data. We also detected some candidate bioindicator foraminiferal species. Based on this proof-of-concept study, we conclude that NGS metabarcoding using foraminifera and other protists has potential to become a new tool for surveying the impact of aquaculture and other industrial activities in the marine environment.