Forum Paper |
Corresponding author: Katy E. Klymus ( kklymus@usgs.gov ) Academic editor: Toshifumi Minamoto
© 2024 Katy E. Klymus, Jacoby D. Baker, Cathryn L. Abbott, Rachel J. Brown, Joseph M. Craine, Zachary Gold, Margaret E. Hunter, Mark D. Johnson, Devin N. Jones, Michelle J. Jungbluth, Sean P. Jungbluth, Yer Lor, Aaron P. Maloy, Christopher M. Merkes, Rachel Noble, Nastassia V. Patin, Adam J. Sepulveda, Stephen F. Spear, Joshua A. Steele, Miwa Takahashi, Alison W. Watts, Susanna Theroux.
This is an open access article distributed under the terms of the CC0 Public Domain Dedication.
Citation:
Klymus KE, Baker JD, Abbott CL, Brown RJ, Craine JM, Gold Z, Hunter ME, Johnson MD, Jones DN, Jungbluth MJ, Jungbluth SP, Lor Y, Maloy AP, Merkes CM, Noble R, Patin NV, Sepulveda AJ, Spear SF, Steele JA, Takahashi M, Watts AW, Theroux S (2024) The MIEM guidelines: Minimum information for reporting of environmental metabarcoding data. Metabarcoding and Metagenomics 8: e128689. https://doi.org/10.3897/mbmg.8.128689
|
Environmental DNA (eDNA) and RNA (eRNA) metabarcoding has become a popular tool for assessing biodiversity from environmental samples, but inconsistent documentation of methods, data and metadata makes results difficult to reproduce and synthesise. A working group of scientists have collaborated to produce a set of minimum reporting guidelines for the constituent steps of metabarcoding workflows, from the physical layout of laboratories through to data archiving. We emphasise how reporting the suite of data and metadata should adhere to findable, accessible, interoperable and reproducible (FAIR) data standards, thereby providing context for evaluating and understanding study results. An overview of the documentation considerations for each workflow step is presented and then summarised in a checklist that can accompany a published study or report. Ensuring workflows are transparent and documented is critical to reproducible research and should allow for more efficient uptake of metabarcoding data into management decision-making.
eDNA, eRNA, metadata, quality control, FAIR data standards, reproducibility
Environmental DNA (eDNA) methods have become increasingly popular in the past two decades due to various advantages compared to traditional methods (
There is momentum in the eDNA community to develop both formal standards (e.g.
There is currently a lack of guidance on metabarcoding data reporting for publication. In order for metabarcoding data to be findable, accessible, interoperable and reproducible (FAIR,
Diagram of an environmental metabarcoding study workflow, including the field (blue), wet laboratory (green) and dry laboratory (orange) components. Created with BioRender.com.
The metabarcoding workflow is similar across a variety of starting sample types often used in biodiversity research, some of which are not typically referred to as eDNA. Here we follow
A checklist for reporting data and metadata associated with environmental metabarcoding studies. Requirements include: ‘Report’ - steps should be reported for FAIR (findable, accessible, interoperable and reproducible) data practices, ‘If applicable’ - report if component is relevant to study, ‘Report for eRNA metabarcoding’ - reporting of additional steps specific to eRNA studies, ‘Optional’ - can be reported. Reporting of these data and metadata will maximise study reproducibility and FAIR data practices. See Suppl. material
Step | Reporting Requirements | Reported |
---|---|---|
Methods – Laboratory Space | ||
General laboratory space layout | Report | |
Contamination mitigation efforts in the laboratory | Report | |
Methods - Metabarcoding Assay | ||
Target gene name | Report | |
Target amplicon length | Report | |
Target taxa | Report | |
Primers, sequence, reference, modifications to published primers | Report | |
Assay validation | Report | |
Methods - Environmental Sample Collection | ||
Number of field samples | Report | |
Definition of field sample | Report | |
Number of field sample replicates | If Applicable | |
Sampling dates | Report | |
Sampling times | If Applicable | |
Sampling locations/Geographic coordinates and geodetic datum used/Non-disclosure statement | Report | |
Capture methods and materials | Report | |
Sample processing method | Report | |
Volume/Mass of sample collected | Report | |
Volume/Mass of sample processed | Report | |
Sample preservation method | Report | |
Sample storage conditions | Report | |
Sample storage duration | Report | |
Environmental parameters | If Applicable | |
Contamination mitigation efforts in the field | Report | |
Positive and Negative (Site/Field/Process) controls | Report | |
Methods - Nucleic Acid Extraction | ||
Extraction method (if using a commercial kit, provide name of kit and manufacturer) | Report | |
Changes/modification to a published method or kit | Report | |
Amount of sample extracted | Report | |
Extraction storage conditions and duration | Report | |
Nucleic acid quantification for each sample and method used to quantify | If Applicable | |
Any subsequent clean-up methods | If Applicable | |
Negative extraction controls | Report | |
Positive extraction controls | If Applicable | |
Methods - Inhibition Detection and Mitigation | ||
PCR inhibition detection and mitigation steps | If Applicable | |
Methods - PCR Amplification and Library Preparation | ||
Library preparation method | Report | |
Number of PCR replicates per sample | If Applicable | |
Are PCR replicates pooled and indexed or indexed separately | If Applicable | |
Thermal cycling instrument and manufacturer | Optional | |
Thermal cycling conditions (temperatures, time at temperature, # cycles, annealing temperature(s)) | Report | |
Master mix composition: Final reaction volume (μl) | Report | |
Master mix composition: name and manufacturer | Report | |
Master mix composition: Taq concentration (X) | Report | |
Master mix composition: Final concentration of each primer (forward and reverse) | Report | |
Master mix composition: Volume of water added (μl) | Report | |
Master mix composition: Volume (μl) or concentration of template NA | Report | |
Master mix composition: Any additives (manufacturers and volumes (μl) or concentrations) | If Applicable | |
Amplicon visualisation: Method used | If Applicable | |
PCR clean-up methods | Report | |
Size selection methods | If Applicable | |
Size selection: Instrumentation and manufacturer | If Applicable | |
Normalisation: Method used | If Applicable | |
PCR controls (Positive PCR, Mock Community, Negative PCR or No-template) and indicate whether or not controls were also sequenced | Report | |
Methods - PCR Amplification and Library Preparation (Reverse Transcription) | ||
Reverse transcriptase reaction kit and manufacturer | Report for eRNA metabarcoding | |
Reverse transcriptase reaction conditions (primers, cycling conditions, controls, amount of template) | Report for eRNA metabarcoding | |
RNA enrichment methods | Report for eRNA metabarcoding | |
DNA contamination evaluation and mitigation methods | Report for eRNA metabarcoding | |
Methods – Sequencing | ||
Sequencing instrument/ platform | Report | |
Sequencing chemistry kit and manufacturer | Report | |
Sequencing quality control steps | Report | |
PhiX and percentage (If using Illumina platform) | If Applicable | |
Methods - Bioinformatics and Reference Database | ||
Database creation: Source of sequences and steps to identify locus of interest | Report | |
Database creation: Method for sequence curation | Report | |
Database creation: Link to database or repository | Report | |
Primer removal (trimming): programme, version, parameters | Report | |
QC programme: programme, version, parameters | Report | |
Read pair merging: programme, version, parameters | Report | |
Chimera removal: programme, version, parameters | Report | |
Clustering: OTUs or ASVs (and thresholds) | Report | |
Additional filtering: removal of singletons or other methods | If Applicable | |
Additional filtering: decontamination using sequenced controls | If Applicable | |
Taxonomic assignment method | Report | |
Taxonomic assignment parameters (and thresholds) | Report | |
Read normalisation: methods | If Applicable | |
Results - Sequencing Summary Statistics | ||
Total number of raw sequence reads produced | Report | |
Total number of reads assigned to MIDs (i.e. tags, indices, barcodes) | Report | |
Total number of reads that made it through bioinformatic filtering | Report | |
Total number of reads used for final/subsequent analyses | Report | |
Total number of OTUs or ASVs assigned to taxa (and to what level of taxonomy) | Report | |
Total number of OTUs or ASVs unassigned | Optional | |
Average number of reads per sample | Optional | |
Minimum and maximum number of reads per sample | Optional | |
Results from Controls (Negative/Positive/Mock Communities) | Report | |
Results - Data Archiving and Availability | ||
Software and code archiving | If Applicable | |
Raw sequence data archiving | Report | |
Processed data archiving | If Applicable |
Several elements should be considered and evaluated before beginning a metabarcoding study (for in-depth review, see
A key consideration is the prevention of contamination when handling eNA samples. This is particularly important in the laboratory setting where PCR is central to library preparation steps. As PCR leads to the amplification of billions of amplicons, samples, reagents, consumables and benchtops can easily become sources of contamination (
The methods sections of published eNA metabarcoding studies should detail specific measures used to prevent laboratory-based contamination, such as: (1) if a unidirectional workflow was used for the wet-laboratory steps (i.e. physically separate laboratory spaces for pre- and post-PCR such that products of later steps are not introduced into spaces from earlier steps); (2) laboratory cleaning protocols and reagents (e.g. sodium hypochlorite solution); (3) workspace designated equipment and consumables; and (4) other laboratory-based contamination prevention measures including positive air pressure, HEPA-filtered air and UV-treatment of workspaces and/or consumables. We acknowledge that not all of these measures will be employed, but we point to ones that should be reported if used. This level of detail is intended to increase transparency and overall confidence in the handling of highly sensitive eNA samples.
Numerous metabarcoding assays (here we define assay as a molecular analysis) have been developed targeting a range of organisms from environmental samples (
Control samples are used throughout the metabarcoding workflow and detailed reporting of all control samples used is vital. Positive controls are generally included to ensure any run or reaction failures or anomalies (i.e. unrelated to the samples themselves) are detected (
Table
Chart of the different elements of the environmental metabarcoding workflow showing the control samples involved at each step.
Table of controls that can be used throughout the metabarcoding workflow. Positive controls confirm a process is working and negative controls ensure any contamination is detected. NA refers to nucleic acid (i.e. DNA or RNA).
Control | How it’s created | Why it’s used |
---|---|---|
Site Positive | Collection of an environmental sample at a site where the target species is known to be present | Confirms assay viability in an environmental sample |
Extraction Positive | A laboratory contrived sample to which NA or tissue is added during the NA extraction process | Introduces target NA to monitor NA extraction efficiency; used for laboratory quality control; not generally used in studies |
Internal Positive Control (IPC) | A known concentration of synthetic or natural NA added to the PCR | Amplification of IPC DNA above expected cycle threshold (qPCR) suggest samples are inhibited; not generally used in metabarcoding, but see text |
PCR Positive | A laboratory contrived sample to which synthetic or natural NA is added during PCR setup; can include mock community samples | Introduces target material to monitor for PCR success |
Site Negative | Collection of an environmental sample at a site where the target species is known to be absent | Monitors potential non-specific amplification of the assay from the environment |
Field Negative | Collection of a blank sample that follows field collection protocols | Monitors potential contamination in field collection |
Process Negative | A sample added during processing that lacks target NA | Monitors potential contamination during sample processing |
Extraction Negative | A laboratory contrived sample that only includes reagents used during the NA extraction process | Monitors potential contamination during extraction process |
PCR Negative (No-template control; NTC) | A laboratory contrived sample that includes only PCR reagents and molecular grade water replaces the NA input | Monitors potential contamination during PCR and sequencing |
Negative Reverse Transcription Control | A sample of RNA extract carried through subsequent steps, but without reverse transcriptase | Tests for DNA contamination in the RNA extract |
Reporting on sample collection for eNA metabarcoding studies allows readers to evaluate the adequacy of the sampling design in the context of the study goals and limitations. A field sample should be defined by the authors. For instance, it could be a singular collection obtained during a single sampling event (i.e. a sample collected at a single location at a single point in time) or a composite sample in which material from multiple locations or times has been combined. Studies may also take replicate field samples given the inevitable stochasticity of whether the target molecule is captured by the sampling if present in the environment. Methods used for capturing samples should be reported, along with sample volume or mass, the number of field samples, field sample replicates and field control samples. In addition, if any pre-extraction processing steps are used after sample collection, the amount of sample processed and methods used (e.g. pre-filtration and/or filtration steps) should be reported. Preservation of samples, storage and duration prior to extraction should be reported as well.
Any contamination mitigation efforts made in the field, including collection of field, site and process controls should also be reported. For recommendations on sample collection and design, see
Detailed information describing the sampling date and locations for all samples should be included. Names of the geographical location of sampling can be reported; however, to ensure reproducibility, geographic coordinates are preferred. The latitude and longitude should be included, as well as the geodetic datum used following Geographic Information Standards (https://www.fgdc.gov/standards). Exceptions for non-disclosure of geographic coordinates are acceptable for privacy of information, cultural concerns or security reasons necessitating exact locations be withheld. Environmental parameters (e.g. temperature, salinity, pH, habitat type) should be reported, if taken, either in a table, supplemental material or within the main text. For further guidance on reporting sampling data, see standards and guidelines developed through Darwin Core (
Extraction protocols are selected based on the sample, tissue or matrix targeted and what steps are required to disrupt biological membranes and release nucleic acids into solution for subsequent purification. Thus, the extraction method should be chosen, based on its efficiency for the particular sample type and/or target taxa and be reported (
Environmental samples are prone to containing molecular compounds and trace metals that can inhibit enzymatic reactions such as PCR, thereby reducing amplification efficiency (
The metabarcoding workflow requires extracted NA samples go through a series of processing steps called library preparation before high-throughput sequencing. Libraries consist of pooled amplicon sequences that have been modified to allow for simultaneous sequencing of high numbers of samples (and subsequent demultiplexing). These modifications include addition of: sequencing primer regions; adaptors compatible to the sequencing platform; and MIDs or multiplex identifiers, that allow identification of sequences to individual samples (
There are multiple methodologies for library preparation and the chosen method should be reported. The three predominant strategies when using Illumina instrumentation are a one-step PCR-based, two-step PCR-based or a ligation-based approach; for a detailed overview of each strategy, see
For all PCR steps in library preparation, chemistries and conditions should be reported. These include: total reaction volume; final concentration of primers; concentrations of master mix components and polymerase; volume of any additives; and amount of template DNA. For mastermix and polymerases, the name and manufacturer should be included. The type of thermal cycler and manufacturer should be reported as well as thermal cycling conditions used. These conditions include the temperature and time at temperature for: the initial denaturation, denaturation, annealing, extension and final extension steps, as well as the number of cycles performed and if annealing temperature varied (e.g. touchdown PCR;
Studies indicate multiple PCR replicates of the same sample are beneficial for environmental metabarcoding studies as they can influence species detection and richness estimates (
Controls should be included and reported in library preparation to assess potential for cross contamination of samples during all PCR steps. Using negative PCR controls allows the user to identify any laboratory contamination and positive PCR controls verify the assay is working. Mock community samples consist of known concentrations of DNA from multiple species and can provide information on amplification efficacy and bias as well as any contamination during library preparation and sequencing (
For eRNA metabarcoding studies, reporting on steps from sample collection through sequencing and data reporting are the same as for eDNA; however, eRNA requires a couple of additional steps in the library preparation phase. Specifically, conversion of RNA to complementary DNA (cDNA) by reverse transcription and post-extraction enrichment of specific RNAs should be reported (
Throughout the library preparation process, sample visualisation is often done to assess the size, distribution and quantity of PCR products, as well as template clean-ups to remove unwanted components. Authors should report methods for DNA product verification as well as where in the process it was done and the type, model and manufacturer of instrumentation used.
One of the primary functions of post-PCR library preparation is to ensure amplified PCR products represent mostly the target sequences; this is done by PCR clean-up and size selection approaches. Size selection can remove non-target DNA products, primer dimer and leftover primer (
A typical last step before sequencing a metabarcoding library is to normalise the concentration of each sample so each is represented equally in the final pooled library sequenced and will have similar read depths (
Sequencing-by-synthesis (SBS) instruments have become the dominant instrumentation for amplicon sequencing (
When using short-read platforms, including Illumina and Element G4 instruments, it is recommended by the manufacturer to spike-in a PhiX sequencing control (Illumina, Inc., San Diego, California, USA; https://singulargenomics.com/g4/reagents/). For ‘low diversity’ libraries (such as those associated with amplicon metabarcoding), PhiX can improve sequencing quality control and provides a measure of overall run performance. If using PhiX, the percentage concentration used should be reported.
If sequencing is outsourced, information from the external facility on what quality control is performed should be obtained before releasing data. For example, most facilities will demultiplex sequence data into sample-specific files and remove sequencing adapter sequences; they may also remove PhiX reads. As with all automated laboratory steps, the instrument platform, sequencing chemistry kit and sequencing quality control steps should be reported.
The process by which raw sequence data are converted to taxon observations and/or counts has many steps, each of which can impact results (e.g.
The comprehensiveness (taxonomic breadth) and curation of reference sequence databases (
Multiple steps are part of the bioinformatic workflow used to convert raw sequence data into biologically analysable data (i.e. a table of sequence read numbers and associated taxonomic assignments). These steps are outlined below and include: Primer removal, Sequence quality control, Read merging, Chimera removal, OTU or ASV creation, Taxonomic assignment, Additional data filtering, and Normalisation of read data. Software workflows that encompass multiple steps exist and include MOTHUR (
Reads produced by sequencing platforms have fixed start positions defined by primers and include the gene region of interest. Failing to remove primers may interfere with taxonomic assignments. Some commonly used programmes for this include CUTADAPT (
After primer removal, sequence read quality controls can be filtered initially based on minimum Q-score values, which are generated by the sequencing instrument for every nucleotide. Bioinformatic programmes like CUTADAPT and TRIMMOMATIC can be used to set quality score thresholds. The quality control programme used and associated parameters should be reported.
When paired-end sequencing is performed, forward and reverse reads can be merged to generate a complete amplicon sequence. Off-target amplification and unremoved primer dimers can result in amplicons with different lengths than expected, which can be bioinformatically filtered out by setting a length threshold. Any reads not passing the threshold will subsequently fail to merge and be removed from the dataset; thus, read type (e.g. forward, reverse, unmerged paired, merged paired), software and parameters used (e.g. minimum number of nucleotide overlap or number of mismatches allowed) should be reported.
Chimeras form during PCR and occur when two different parent DNA strands anneal together to create PCR artefacts that do not exist in nature. Chimeras can be difficult to identify (
Operational taxonomic units (OTUs) or amplicon sequence variants (ASVs) are created from sequencing reads, which reduces the overall size of datasets and computational power needed to analyse them (
There are many methods to assign taxonomy to OTUs/ASVs. The most well-known and longest supported method is the sequence similarity-based top BLAST hit approach (
Taxonomic assignment of some reads may not be resolvable to species level for a number of reasons. When percentage identity parameters fall below a user-defined threshold or if the reference database is sparsely populated, a single OTU/ASV might be assigned to multiple species. Furthermore, higher taxonomic levels such as genus or family may be targeted for assignment rather than at the species level. When reporting on taxonomy assignment, include methods used, threshold parameters, the reference database(s) and the taxonomic ranks assigned to each OTU/ASV. For more information on taxonomic assignment, we refer readers to
Bioinformatic decontamination, not to be confused with laboratory decontamination, conducts post-processing quality control of OTU/ASV tables. No consensus exists on how to filter data using sequenced PCR controls. This process may employ a range of steps, including cut-offs based on sample sequencing read depth, OTU/ASV prevalence across samples, OTU/ASV proportional abundance and reads identified in the positive and negative PCR controls (
Normalisation of read data is the process by which reads are scaled or transformed to allow more accurate comparisons across samples. Normalising high throughput sequencing read data can help account for biases that occur during PCR and sequencing (
Initial summary metrics from a sequencing run are often unreported, but are helpful to evaluate sequencing efficiency and evaluate metabarcoding results. They also indicate whether library preparation methods were highly effective or if changes might be necessary for future studies. Relevant summary metrics to report are: total number of raw reads; number or percentage of reads assigned to MIDs; total number or percentage of reads that passed bioinformatic filter thresholds (i.e. quality control, trimming and merging); number or percentage of reads assigned to taxonomy; the taxonomic level of assignments (i.e. species level, family level, phyla level); total number or percentage of reads unassigned to taxonomy; and total number of reads used in the final analysis or any subsequent analyses. Additionally, reporting per-sample metrics such as average number of reads per sample and minimum and maximum number of reads per sample allows an evaluation of both sequencing depth and evenness across samples.
The practice of sequencing control samples (both negative and positive) is highly recommended in the literature as deviations from expected results may indicate issues that should be addressed or accounted for either computationally or in data interpretation (
Alongside peer-review publications, data sharing is a core pillar of the sciences. Raw sequence data should be deposited in the International Nucleotide Sequence Database Collaboration (INSDC) nucleotide sequence archives (e.g. NCBI, ENA, DDBJ). Data and metadata should adhere to established standards such as the Minimum information about a marker gene sequence (MIMARKS) and minimum information about any sequence specifications (MixS) developed by Genomic Standards Consortium (
Although significant bottlenecks to achieving completely FAIR metabarcoding data practices remain, at minimum, all data (i.e. OTU/ASV table, sequencing data, metadata), methods and code used to generate, analyse and interpret metabarcoding data should be provided to open-access repositories to support open science principles and enhance trust and reproducibility. Metabarcoding studies generate valuable biodiversity data and FAIR practices will allow biodiversity monitoring at increased speeds and scales, which is needed given current global biodiversity loss. All efforts should be made to adhere to the growing consensus to serve biodiversity data in large international biodiversity repositories (e.g. Global Biodiversity Information Facility (GBIF Data Standards) and Ocean Biodiversity Information System (OBIS)). In particular, we point users to
The advent of DNA metabarcoding has transformed our ability to census and assess biological communities. With this new capacity for generating biological data at increasing sensitivity and scale comes a deluge in environmental DNA research datasets, hence it is important that we pause and take stock of what minimum metadata should accompany environmental metabarcoding publications. Here, we identified a suite of sampling, analytical and data archiving information that should be included in publications to meet FAIR data standards and provide context for eNA results to be repeatable and interpretable. We recommend authors report these in the manuscript, supplemental materials or online resources linked to the publication (e.g. GitHub, protocols.io. etc.). This is crucial for the use and reuse of eNA data in global scale biomonitoring efforts (
We recognise that the generation and curation of metabarcoding data is time and labour-intensive and that analyses require substantial computational resources and bioinformatic expertise. This can severely limit the ability of the metabarcoding community to process data quickly and efficiently into actionable biodiversity information (
The authors thank their respective research groups, collaborators and the eDNA community whose collective experiences led to the idea for this manuscript. We also thank Dr. Freya Rowland for assistance with figures and Dr. Brittany Perrotta for helpful comments on this paper. Any use of trade, firm or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the U.S. Fish and Wildlife Service.This publication represents NOAA Pacific Marine Environmental Laboratory Contribution No. 5641.
The authors have declared that no competing interests exist.
No ethical statement was reported.
No funding was reported.
Conceptualization: Katy Klymus, Zachary Gold, Susanna Theroux. Visualization: Jacoby Baker, Katy Klymus. Writing- original draft: Katy Klymus, Jacoby Baker, Cathryn Abbott, Rachel Brown, Joseph Craine, Zachary Gold, Margaret Hunter, Mark Johnson, Devin Jones, Michelle Jungbluth, Sean Jungbluth, Yer Lor, Aaron Maloy, Christopher Merkes, Rachel Noble, Nastassia Patin, Adam J Sepulveda, Stephen Spear, Joshua Steele, Miwa Takahashi, Alison Watts, Susanna Theroux. Writing- review & editing: Susanna Theroux, Cathryn Abbott, Katy Klymus. Supervision/ Admisitration: Katy Klymus, Susanna Theroux.
Katy E. Klymus https://orcid.org/0000-0002-8843-6241
Jacoby D. Baker https://orcid.org/0000-0002-0673-7535
Cathryn L. Abbott https://orcid.org/0000-0002-5314-7351
Rachel J. Brown https://orcid.org/0000-0001-5353-715X
Joseph M. Craine https://orcid.org/0000-0001-6561-3244
Zachary Gold https://orcid.org/0000-0003-0490-7630
Margaret E. Hunter https://orcid.org/0000-0002-4760-9302
Mark D. Johnson https://orcid.org/0000-0002-0460-945X
Devin N. Jones https://orcid.org/0000-0001-9215-2930
Michelle J. Jungbluth https://orcid.org/0000-0001-9339-7497
Sean P. Jungbluth https://orcid.org/0000-0001-9265-8341
Yer Lor https://orcid.org/0000-0002-5738-2412
Aaron P. Maloy https://orcid.org/0000-0003-4412-2280
Christopher M. Merkes https://orcid.org/0000-0001-8191-627X
Rachel Noble https://orcid.org/0000-0001-9071-8312
Nastassia V. Patin https://orcid.org/0000-0001-8522-7682
Adam J. Sepulveda https://orcid.org/0000-0001-7621-7028
Stephen F. Spear https://orcid.org/0000-0001-8351-9382
Joshua A. Steele https://orcid.org/0000-0001-8023-8956
Miwa Takahashi https://orcid.org/0000-0001-8952-051X
Alison W. Watts https://orcid.org/0000-0001-9700-6393
Susanna Theroux https://orcid.org/0000-0002-9812-7856
All of the data that support the findings of this study are available in the main text or Supplementary Information.
Additional information
Data type: xlsx
Explanation note: file S1. A printable checklist of data and metadata that should be reported or reported if available for environmental metabarcoding studies. file S2. The checklist of data and metadata that should be reported or reported if available for environmental metabarcoding studies, including example entries for reference.