Metagenomics and Environmental DNA (eDNA) Analysis: Principles, Workflows, and Veterinary Applications

Metagenomics and environmental DNA (eDNA) analysis have transformed the capacity to detect and characterize biological communities without direct observation or capture of organisms. In veterinary medicine, these methods enable surveillance of pathogens across livestock, companion animals, wildlife, and aquatic species through non-invasive sampling of water, soil, air, feed, or host-associated matrices. This review synthesizes the molecular principles, bioinformatic pipelines, and diagnostic applications of eDNA metagenomics with emphasis on veterinary contexts, incorporating recent methodological advances.

Principles of eDNA and Metagenomics

Environmental DNA refers to genetic material shed by organisms into surrounding environments through sloughed cells, secretions, feces, gametes, or decomposing tissues. eDNA can exist as intracellular (within intact cells) or extracellular (free or adsorbed to particles) DNA, with persistence influenced by temperature, pH, UV exposure, microbial activity, and adsorption to sediment [1, 2]. Metagenomics involves the direct sequencing of total DNA extracted from an environmental sample, encompassing all genomes present. Two principal strategies exist:

1. Targeted metabarcoding: PCR amplification of a conserved genetic marker (e.g., 16S rRNA for bacteria, 18S for eukaryotes, cytochrome c oxidase I for animals, or internal transcribed spacer for fungi) followed by high-throughput sequencing. This approach trades taxonomic breadth for depth and is cost-effective for known marker regions.

2. Shotgun metagenomics: Non-targeted sequencing of total DNA, yielding fragments from all organisms including viruses, prokaryotes, and eukaryotes. This approach enables functional gene annotation, pathogen discovery, and strain-level resolution but requires greater sequencing depth and computational resources.

The choice between strategies depends on the ecological question, target organism group, and resource constraints. Workflow trade-offs have been systematically evaluated using transparent frameworks applicable to resource-limited veterinary laboratories [3].

Sample Collection, Preservation, and DNA Extraction

Optimal eDNA recovery begins with collection strategy. Water samples typically require filtration (0.22 to 5.0 m membrane filters) or precipitation (e.g., ethanol or cetyltrimethylammonium bromide methods). Sediment, soil, feces, and biofilm samples require homogenization and removal of inhibitory compounds such as humic acids and polysaccharides. Preservation is critical: samples should be stored at 4 degrees Celsius for short term or frozen at -20 degrees Celsius or -80 degrees Celsius; ethanol preservation is an alternative for field conditions.

DNA extraction from environmental matrices must balance yield with purity. Commercial silica column kits and magnetic bead-based systems are widely used, but bead-beating steps are essential for lysing microbial cells. A critical methodological nuance is the carryover of residual eDNA into RNA extracts. Wang et al. [2] demonstrated that eDNA contamination in eRNA extracts significantly skews biodiversity assessments, producing false positives for taxa not transcriptionally active. They advocate for rigorous DNase treatment protocols to ensure that eRNA-based community profiles accurately reflect the expressed metatranscriptome. This finding is particularly relevant for veterinary studies aiming to differentiate viable or active pathogens from relic DNA.

Amplification and Sequencing Strategies

For metabarcoding, primer selection defines the detectable taxonomic range. Bacterial communities are commonly profiled using primers targeting the V3-V4 or V4 regions of the 16S rRNA gene. Fungal diversity is assessed via ITS1 or ITS2 regions [1]. Eukaryotic metazoans and protists are increasingly surveyed using 18S rRNA V9 or 12S/16S mtDNA primers. Morissette et al. [4] applied eDNA metabarcoding with 12S primers to assess anthropogenic disturbance effects on freshwater fish communities, demonstrating that trait-based biomonitoring indices derived from eDNA accurately reflect ecological degradation.

Multiplexing is achieved through indexing PCR, and sequencing is performed on high-throughput platforms (generic term for instruments that generate millions of short reads). For shotgun metagenomics, library preparation involves DNA fragmentation, end repair, adapter ligation, and optional target enrichment (e.g., probe-based capture of viral genomes).

Bioinformatic Analysis Pipelines

Raw sequencing data undergo quality filtering (removal of adapter contamination, low-quality bases, and chimeric sequences). For metabarcoding, operational taxonomic units (OTUs) or amplicon sequence variants (ASVs) are generated and assigned taxonomy using reference databases (e.g., Silva for bacteria, PR2 for protists, UNITE for fungi). Shotgun data require assembly of reads into contigs, gene prediction, and functional annotation. Taxonomic binning can be performed using k-mer-based classifiers or alignment-based methods.

A major challenge is detecting hidden diversity, particularly among microbial eukaryotes. Geerts et al. [5] introduced disentangled assembly graphs, a computational approach that resolves complex metagenomic assemblies to recover eukaryotic genomes that are often missed by standard pipelines. This method uses graph-based decomposition to separate co-assembled sequences from different organisms, thereby improving taxonomic assignment and genome recovery. For veterinary parasitology, such tools are invaluable for resolving mixed infections of helminths, protozoa, and fungi from environmental or fecal samples.

Applications in Veterinary Medicine

Pathogen Surveillance in Livestock and Poultry

eDNA metagenomics can detect pathogens in barn air, water, feed, and bedding without animal handling. For example, airborne bacterial and viral genomes can be captured using filter-based air samplers and sequenced to identify agents ofrespiratory disease such as Mycoplasma bovis in cattle, Mannheimia haemolytica, and Clostridium perfringens types. In poultry houses, eDNA from dust and litter can reveal the presence of Escherichia coli strains, Salmonella serovars, and parasites such as Eimeria spp. and Histomonas meleagridis. Fecal eDNA from communal water sources can serve as a pooled surveillance tool for enteric pathogens like Salmonella in chickens or Lawsonia intracellularis in swine.

Aquatic Animal Health and Aquaculture

eDNA methods are particularly suited for aquatic environments because water integrates genetic material from all organisms present. In aquaculture settings, eDNA can monitor waterborne pathogens such as Streptococcus agalactiae and Streptococcus iniae in tilapia, Ichthyophthirius multifiliis in fish, and sea lice (Lepeophtheirus salmonis) in salmon. The approach allows early detection before clinical outbreaks, enabling timely intervention. Long et al. [6] used eDNA metagenomics to explore interactions between phytoplankton and bacterioplankton communities in a plateau lake; similar framework can be applied to shrimp ponds or recirculating aquaculture systems to predict dysbiosis preceding mortality.

Wildlife Disease Ecology and Zoonotic Risk Assessment

eDNA from water, soil, or feces can detect pathogens shed by wildlife, including Pasteurella multocida (avian cholera), Francisella tularensis (tularemia), Mycobacterium avium subsp. paratuberculosis, and tick-borne parasites such as Babesia and Theileria. Predator-prey dynamics can be inferred from eDNA: Sato et al. [7] determined predator composition in rivers using eDNA analyses of water combined with color pattern assessments of guppies, illustrating how eDNA can trace trophic interactions and pathogen transmission routes. eDNA surveillance at wildlife-livestock interfaces is critical for notifiable diseases such as highly pathogenic avian influenza (H5N1) and African swine fever.

Parasite Detection and Biomonitoring

Parasites often have complex life cycles involving multiple hosts and environmental stages. eDNA can detect free-living stages (eggs, larvae, sporocysts) in water, soil, or vegetation. For example, Fasciola hepatica eggs in cattle pasture runoff, Teladorsagia circumcincta larvae on herbage, and Dicrocoelium dendriticum eggs in ant nests are all accessible via eDNA. Metabarcoding of soil and water can also capture intermediate hosts such as snails for Fasciola or blackflies for Leucocytozoon.

Seasonal and Resilience Monitoring

Adedire et al. [1] applied ITS2 metabarcoding to assess fungal phylotype diversity across seasons in a tropical spring, demonstrating snapshot resilience of fungal communities. In veterinary contexts, such temporal monitoring can track seasonal emergence of pathogenic fungi like Macrorhabdus ornithogaster in birds or Aspergillus species in poultry litter. Repeat eDNA sampling along environmental gradients can reveal how climate variables shape pathogen reservoirs.

Methodological Limitations and Quality Control

Despite its power, eDNA metagenomics carries several limitations that veterinary diagnosticians must consider:

• Degradation and false negatives: eDNA degrades rapidly in warm, UV-exposed, or microbially active environments. Sampling must be timed to match peak shedding and environmental persistence.
• PCR bias: Metabarcoding primers preferentially amplify certain taxa due to mismatches or differential copy numbers of marker genes, leading to skewed relative abundances.
• Reference database gaps: Many veterinary pathogens, particularly novel or poorly characterized viruses and parasites, lack reference sequences. Shotgun metagenomics with de novo assembly partially addresses this, but binning tools may miss divergent genomes. Disentangled assembly graphs [5] offer a solution for eukaryotic microbes.
• Residual eDNA in RNA extracts: As highlighted by Wang et al. [2], co-extracted DNA in RNA samples inflates apparent transcriptional activity. DNase treatment must be validated to avoid misinterpretation of viability.
• Resource constraints: Laboratories in low-resource settings require flexible workflows. Ip et al. [3] provided a transparent framework that evaluates trade-offs in cost, time, detection sensitivity, and taxonomic resolution, facilitating protocol optimization for field conditions.

Standardized Workflow and Decision Framework

The following Mermaid diagram outlines a decision tree for selecting an eDNA metagenomics approach in a veterinary diagnostic context.

flowchart TD
    A[Define target organisms and questions], > B{Pathogen of interest known?}
    B, >|Yes| C[Select marker gene for metabarcoding]
    B, >|No| D[Perform shotgun metagenomics]
    C, > E[Primer design and validation]
    E, > F[Sample collection and filtration]
    D, > F
    F, > G[DNA extraction and purification]
    G, > H[QC: quantify and assess purity]
    H, > I{Meets QC thresholds?}
    I, >|No| J[Re-extract or concentrate]
    I, >|Yes| K[Library preparation and indexing]
    K, > L[High-throughput sequencing]
    L, > M[Data preprocessing: trim, filter, de-noise]
    M, > N[Metabarcoding: OTU/ASV assignment]
    M, > O[Shotgun: assembly, binning, annotation]
    N, > P[Taxonomic profiling]
    O, > P
    P, > Q[Interpretation: pathogen detection, diversity indices, functional potential]
    Q, > R[Validation via qPCR or culture if indicated]

Comparative Table of eDNA Approaches

Integration with Existing Veterinary Diagnostic Infrastructure

eDNA metagenomics complements rather than replaces traditional diagnostics such as culture, serology (e.g., ELISA for p27 antigen detection in feline leukemia virus), and targeted PCR. It excels in scenarios where multiple pathogens are suspected, hosts are difficult to sample, or environmental reservoirs must be characterized. For example, in investigating an outbreak of necrotic enteritis in broilers, eDNA from litter and water can simultaneously detect Clostridium perfringens toxin genes, Eimeria oocysts, and dysbiosis markers. Likewise, in ruminant flocks, eDNA from pasture can map the distribution of Haemonchus contortus and Nematodirus battus larvae.

The approach also supports one health surveillance by linking environmental, animal, and human pathogens. However, veterinary professionals must be trained in bioinformatic interpretation to avoid false positives due to contamination or database errors. Cross-listing of detected sequences with curated veterinary pathogen databases is essential.

References

[1] Adedire DE, Onilude AA, Odeniyi OA, et al. Snapshot reflection of the seasonal resilience and diversity of fungal phylotypes in the tropical Ikogosi spring. Environ Sci Pollut Res Int. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42135536/

[2] Wang F, Xiong W, Huang X, et al. Residual eDNA in eRNA Extracts Skews eRNA-Based Biodiversity Assessment: Call for Optimised DNase Treatment. Mol Ecol Resour. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41552860/

[3] Ip YCA, Allan EA, Hirsch SL, et al. Fast, Flexible, Feasible: A Transparent Framework for Evaluating eDNA Workflow Trade-Offs in Resource-Limited Settings. Mol Ecol Resour. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41482808/

[4] Morissette O, Côté G, Couillard MA, et al. Trait-Based Biomonitoring Using eDNA Metabarcoding to Assess Anthropogenic Disturbances on Freshwater Fish Communities. Mol Ecol Resour. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41858251/

[5] Geerts MM, Curto M, Alverson AJ, et al. Disentangled Assembly Graphs Reveal Hidden Eukaryotic Diversity in eDNA Metagenomic Data. Mol Ecol Resour. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41858257/

[6] Long Y, Guo J, Dai L, et al. Interactions between phytoplankton and bacterioplankton communities in Caohai plateau lake, revealed by environmental DNA metagenomics. BMC Microbiol. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41491676/

[7] Sato Y, Sato Y, Deki O, et al. Estimated predator composition using environmental DNA analyses and color patterns of male guppies in introduced rivers. Sci Rep. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41495321/