Metagenomic Next-Generation Sequencing (mNGS) for Veterinary Diagnostics: Challenges and Applications

Introduction

Metagenomic next-generation sequencing (mNGS) represents a paradigm shift in veterinary diagnostic microbiology, moving from targeted pathogen detection to an unbiased, hypothesis-free approach capable of identifying all nucleic acids present in a clinical sample [1]. Unlike traditional methods such as culture, polymerase chain reaction (PCR), or serology, mNGS does not require a priori knowledge of the pathogen's identity [2]. This capability is particularly valuable in veterinary medicine, where novel, emerging, or unexpected pathogens frequently cause disease outbreaks in livestock, companion animals, and wildlife [3, 4, 5]. The technique simultaneously detects viruses, bacteria, fungi, and parasites, providing a comprehensive view of the microbial community and enabling the discovery of co-infections that may be missed by targeted assays [6, 7]. This article provides a detailed technical review of mNGS workflows, bioinformatics analysis, diagnostic applications, and inherent challenges within the veterinary context.

The mNGS Workflow: From Sample to Sequence

The mNGS diagnostic pipeline comprises several critical stages: sample collection and preparation, nucleic acid extraction, library construction, high-throughput sequencing, and bioinformatic analysis [8, 9]. Each step introduces potential biases that can affect the sensitivity and specificity of the final result.

Sample Collection and Pre-Processing

The choice of sample type is dictated by the clinical syndrome. For respiratory disease, nasal swabs, bronchoalveolar lavage fluid (BALF), or lung tissue are common [10, 11, 12]. For enteric disease, fecal samples or intestinal contents are used [5, 13, 7]. For systemic or neurological cases, whole blood, serum, or cerebrospinal fluid may be analyzed [14]. The high background of host nucleic acid, particularly in tissue samples, is a major challenge [2]. To enrich for microbial sequences, pre-processing steps such as centrifugation, filtration (e.g., 0.45 or 0.22 micron filters to remove host cells and large debris), or nuclease treatment (e.g., DNase/RNase digestion of free host nucleic acids) are often employed, especially for viral metagenomics [2, 1]. For RNA virus detection, samples must be preserved in RNA-stabilizing solutions or processed rapidly to prevent degradation [9].

Nucleic Acid Extraction and Library Preparation

Total nucleic acid (DNA and RNA) is extracted using column-based or bead-based methods [9]. For RNA virus detection, a reverse transcription step is required to generate complementary DNA (cDNA) [15]. The extracted nucleic acid is then fragmented, and adapter sequences are ligated to the ends to create a sequencing library [8]. A critical step is the choice between total RNA sequencing (ribodepletion) and poly-A selection. Ribodepletion is preferred for metagenomics as it removes ribosomal RNA (rRNA) from both host and microbes, enriching for messenger RNA and non-coding RNA without biasing against non-polyadenylated viral genomes [2, 1]. Probe-based enrichment (hybridization capture) can be used to increase sensitivity for specific pathogen groups by using biotinylated oligonucleotide probes to pull down target nucleic acids from the library [16, 17, 11]. This approach balances the breadth of metagenomics with the sensitivity of targeted sequencing [16].

Sequencing Platforms

High-throughput sequencing platforms are central to mNGS. Short-read sequencers produce millions of reads (typically 75-300 base pairs) with high accuracy, making them suitable for taxonomic classification and genome assembly [8]. Long-read sequencers generate reads of several kilobases, which are advantageous for resolving repetitive regions, identifying structural variants, and assembling complete viral genomes in real time [8]. The choice of platform depends on the required turnaround time, read length, throughput, and cost. A detailed comparison of these technologies is available in the article on Long-Read Sequencing Technologies: PacBio and Oxford Nanopore.

Bioinformatics Analysis: From Raw Reads to Pathogen Identification

Bioinformatics is the most complex and critical component of the mNGS workflow [18]. The primary goal is to distinguish pathogen-derived sequences from the overwhelming background of host and commensal microbial nucleic acids.

Quality Control and Host Read Depletion

Raw sequencing reads undergo quality control to remove adapter sequences, low-quality bases, and short reads [18]. High-quality reads are then aligned to the host reference genome (e.g., the relevant animal species) using rapid aligners. Reads that map to the host are discarded, leaving a subset of non-host reads for further analysis [2, 18]. This step is computationally intensive but essential for reducing data volume and improving detection sensitivity.

Taxonomic Classification

The non-host reads are classified by comparison against comprehensive nucleotide or protein databases. Two main algorithmic approaches are used:

1. Alignment-based methods: Tools like BLAST or DIAMOND align reads against reference databases (e.g., NCBI GenBank, RefSeq). This is sensitive but computationally slow.
2. K-mer based methods: Tools like Kraken2 classify reads by matching short subsequences (k-mers) against a pre-built database of unique k-mers from known genomes [18]. This is extremely fast and suitable for real-time analysis.

Pipelines such as TaxTriage integrate multiple tools to provide a robust classification, filtering out common environmental contaminants and commensals [18]. The output is a list of putative pathogens with associated read counts, coverage depth, and taxonomic ranks.

Genome Assembly and Interpretation

For novel or highly divergent pathogens, read-based classification may fail. In such cases, de novo assembly is used to reconstruct longer contiguous sequences (contigs) from overlapping reads [2, 4]. These contigs can be compared to viral or bacterial protein databases to identify distant homologs. The detection of a pathogen is confirmed by assessing the breadth and depth of genome coverage, the number of unique reads mapping to the pathogen, and the absence of the agent in negative controls [2, 19]. The interpretation must consider the clinical context, as the mere presence of a potential pathogen does not prove causation, particularly for opportunistic agents or those found in the normal microbiome [20].

flowchart TD
    A[Clinical Sample<br>Blood, Swab, Feces, Tissue], > B[Pre-Processing<br>Filtration, Nuclease Treatment]
    B, > C[Nucleic Acid Extraction<br>DNA/RNA]
    C, > D{Target Enrichment?}
    D, No, > E[Library Preparation<br>Fragmentation, Adapter Ligation]
    D, Yes, > F[Hybridization Capture<br>Probe-based Enrichment]
    F, > E
    E, > G[High-Throughput Sequencing<br>Short-read or Long-read]
    G, > H[Raw Reads]
    H, > I[Quality Control<br>Adapter Trimming, Filtering]
    I, > J[Host Read Depletion<br>Alignment to Host Genome]
    J, > K[Non-Host Reads]
    K, > L{Taxonomic Classification}
    L, > M[Alignment-based<br>BLAST/DIAMOND]
    L, > N[K-mer based<br>Kraken2]
    M, > O[Pathogen List<br>Read Counts, Coverage]
    N, > O
    O, > P{Novel Pathogen?}
    P, No, > Q[Interpretation<br>Clinical Correlation]
    P, Yes, > R[De Novo Assembly]
    R, > S[Contig Analysis<br>Protein Homology Search]
    S, > Q
    Q, > T[Diagnostic Report]

Applications in Veterinary Diagnostics

mNGS has been applied across a wide range of veterinary scenarios, demonstrating its utility in outbreak investigations, pathogen discovery, and complex diagnostic cases.

Pathogen Discovery and Outbreak Investigation

mNGS is uniquely suited for identifying the etiological agent in disease outbreaks where conventional testing is negative. In a retrospective analysis of a mortality event in wild black vultures, mNGS led to the discovery of a novel bandavirus [4]. Similarly, metagenomic sequencing identified a novel caprine astrovirus in a goat kid with neurological disease [21] and a potentially undescribed columnaris-causing bacterium in rainbow trout [22]. In swine, mNGS has revealed the complexity of viral communities associated with diarrhea, identifying co-infections with multiple viruses including rotaviruses, coronaviruses, and picornaviruses [5, 7]. The technique has also been used to characterize the virome of quail farms experiencing severe intestinal disease, implicating deltacoronaviruses and picornaviruses [23].

Respiratory Disease Complex

Bovine respiratory disease (BRD) and porcine respiratory disease complex (PRDC) are multifactorial, involving viral and bacterial pathogens. mNGS provides a comprehensive snapshot of the entire respiratory microbiome. Hybridization capture sequencing has been used to enhance the detection of BRD pathogens and antimicrobial resistance genes directly from clinical samples [11]. In horses, mNGS has identified viral pathogens in respiratory cases, including equine herpesviruses and influenza A virus [12]. The ability to detect unexpected agents, such as Rickettsia felis causing pneumonia in a human case diagnosed by targeted NGS, highlights the potential for mNGS to identify zoonotic pathogens in animals [10].

Emerging and Zoonotic Pathogen Surveillance

The One Health concept emphasizes the interconnectedness of human, animal, and environmental health. mNGS is a cornerstone technology for viral surveillance at the animal-human interface [2, 1]. It has been used to characterize the genetic diversity of pathogens like porcine reproductive and respiratory syndrome virus (PRRSV) in live virus inoculum [6] and to identify tick-carried pathogens on deer [24]. The technique is critical for monitoring the emergence of novel influenza strains, coronaviruses, and other zoonotic agents in animal reservoirs [1, 25]. For a deeper dive into aquatic systems, see Metagenomic Sequencing for Aquatic Viral Pathogens.

Complex and Atypical Cases

When standard diagnostics fail, mNGS can provide a definitive diagnosis. It has been used to diagnose paralytic rabies in a dog, where traditional methods were inconclusive [14]. In a case of severe panophthalmitis in a dog, mNGS of formalin-fixed, paraffin-embedded tissue identified an Actinomyces species, a fastidious bacterium difficult to culture [26]. The technique has also been applied to diagnose elephant endotheliotropic herpesvirus infection in Asian elephants [27] and to characterize a novel Castellaniella species in dogs [28].

Comparison with Traditional Diagnostic Methods

mNGS offers several advantages over traditional methods but also has distinct limitations.

| Feature | mNGS | Targeted PCR / qPCR | Culture | | :-, | :-, | :-, | :-, | | Scope | Unbiased, pan-pathogen | Targeted, known agents | Targeted, viable agents | | Sensitivity | Moderate to high (depends on host background) | Very high | Low to moderate | | Specificity | High (with bioinformatics filters) | Very high | High (for viable organisms) | | Turnaround Time | 24-72 hours | 2-6 hours | Days to weeks | | Cost per Sample | High | Low to moderate | Low to moderate | | Novel Pathogen Detection | Yes | No | Possible but difficult | | Quantification | Semi-quantitative (relative abundance) | Quantitative (Ct value) | Semi-quantitative (CFU) | | Bioinformatics Need | Extensive | Minimal | None |

While PCR remains the gold standard for rapid, sensitive, and cost-effective detection of known pathogens [13], mNGS excels in scenarios requiring broad discovery. Culture is essential for antimicrobial susceptibility testing and strain isolation but is slow and fails for many fastidious organisms. mNGS can detect non-viable or difficult-to-culture pathogens, such as viruses and intracellular bacteria [26, 14]. For absolute quantification of viral load, techniques like digital droplet PCR (ddPCR) offer superior precision, as discussed in Digital Droplet PCR (ddPCR) for Absolute Quantification of Viral Load in Veterinary Diagnostics: Principles and Applications.

Challenges and Limitations

Despite its power, the routine implementation of mNGS in veterinary diagnostics faces several significant hurdles.

Sensitivity and Host Background

The sensitivity of mNGS is inversely proportional to the amount of host nucleic acid in the sample [2]. In tissue biopsies, host DNA can constitute over 99% of the total sequenced reads, leaving very few reads for pathogen detection. This can lead to false negatives, especially for low-titer infections. Enrichment strategies, such as hybridization capture, can improve sensitivity by 10 to 100-fold for target pathogens [16, 17, 11].

Contamination and Interpretation

Environmental and reagent contamination is a major source of false positives [19]. DNA from laboratory reagents, water, or the environment can be co-sequenced and misidentified as a pathogen. Rigorous use of negative controls (e.g., nuclease-free water processed alongside samples) and bioinformatic filtering of common contaminants (e.g., Pseudomonas, Escherichia coli) are essential [18, 19]. Interpreting the clinical significance of detected organisms, particularly commensals or opportunists, requires expert knowledge and correlation with clinical and histopathological findings [20].

Cost and Turnaround Time

The cost of sequencing and bioinformatics analysis remains higher than that of targeted PCR panels [8]. While sequencing costs have decreased, the total cost per sample, including library preparation and analysis, is still a barrier for many veterinary laboratories. Turnaround time, typically 24-72 hours, is longer than for point-of-care tests or real-time PCR, which can be critical for acute disease management [8].

Bioinformatics Complexity and Standardization

The bioinformatics pipeline is not standardized, leading to variability in results between laboratories [19]. An interlaboratory proficiency test using mNGS for RNA virus detection in swine feces showed that while all laboratories correctly identified the spiked virus, there was significant variation in the reported number of reads and the detection of other background viruses [19]. The need for specialized computational infrastructure and expertise is a major limitation for many diagnostic labs. For a detailed guide on NGS workflows, see From Raw Reads to Variants: A Diagnostic Blueprint for Next-Generation Sequencing (NGS) Workflows.

RNA Virus Detection

RNA viruses are more labile than DNA viruses, and their genomes are susceptible to degradation during sample collection, transport, and processing [9]. The efficiency of reverse transcription and second-strand synthesis can also introduce bias. Optimized protocols for RNA extraction and library preparation are critical for reliable RNA virus detection [15, 9].

Future Directions

The future of mNGS in veterinary diagnostics lies in automation, miniaturization, and integration with other technologies. The development of portable long-read sequencers is enabling real-time genomic surveillance in the field, such as for avian influenza in poultry [8]. Advances in machine learning and artificial intelligence are improving the accuracy and speed of taxonomic classification and the prediction of pathogenicity from genomic data. The integration of mNGS with other diagnostic modalities, such as serology and digital pathology, will provide a more holistic view of the disease state. Furthermore, the establishment of standardized protocols, reference materials, and proficiency testing programs is essential for the widespread clinical adoption of mNGS in veterinary medicine [19].

Conclusion

Metagenomic next-generation sequencing is a transformative technology for veterinary diagnostics, offering an unbiased, comprehensive approach to pathogen detection and discovery. Its applications in outbreak investigations, complex disease cases, and One Health surveillance are well documented. However, challenges related to sensitivity, cost, contamination, and bioinformatics standardization must be addressed before mNGS can become a routine first-line diagnostic tool. As the technology matures and becomes more accessible, it will play an increasingly vital role in safeguarding animal health and understanding the complex microbial ecology of disease.

References

[1] Russell T, Formiconi E, Casey M et al. Viral Metagenomic Next-Generation Sequencing for One Health Discovery and Surveillance of (Re)Emerging Viruses: A Deep Review. Int J Mol Sci. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/41097095/

[2] Russell T, Formiconi E, Murphy A et al. One health viral metagenomics for pathogen surveillance: robust mNGS workflows for viral detection and genome recovery from swab and tissue specimens. BMC Microbiol. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42215894/

[3] Kariithi HM, Volkening JD, Mueni SN et al. Co-detection and genomic characterization of avian rotavirus A, avian orthoreovirus, and chicken megrivirus-C using nontargeted metagenomic surveillance in Indian broiler chickens. Front Cell Infect Microbiol. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41835009/

[4] Mwakibete L, Hoarau AOG, Ahyong V et al. Discovery of a novel bandavirus using metagenomic sequencing in a retrospective analysis of an unresolved 2020 mortality event involving wild black vultures in the northeastern United States. J Vet Diagn Invest. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41685543/

[5] Maan S, Batra K, Rajendhran J et al. Exploring viral diversity in diarrheic porcine feces: a metagenomic analysis from an Indian swine farm. Front Cell Infect Microbiol. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/41017909/

[6] Mohammed MZ, Linhares DCL, Zeller MA et al. Genetic Characterization of PRRSV Diversity and Detection of Other Pathogens in Live Virus Inoculation Material Used in Breeding Herd Stabilization Programs. Microorganisms. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42354835/

[7] Li Q, Jiang J, Li J et al. Complexity of Diarrhea-Associated Viruses in Stunted Pigs Identified by Viral Metagenomics. Transbound Emerg Dis. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40567259/

[8] Chaves M, Hashish A, Goraichuk IV et al. Nanopore sequencing in veterinary medicine: from concepts to clinical applications. Front Cell Infect Microbiol. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/41341958/

[9] Chaves M, Hashish A, Osemeke O et al. Evaluation of Commercial RNA Extraction Protocols for Avian Influenza Virus Using Nanopore Metagenomic Sequencing. Viruses. 2024. URL: https://pubmed.ncbi.nlm.nih.gov/39339905/

[10] Bai Y, Zhao J, Wang Z et al. A case of Rickettsia felis caused pneumonia and diagnosed by clinical analysis and Targeted Next-Generation Sequencing (tNGS) using Bronchoalveolar Lavage Fluid (BALF): A case report and literature review. J Infect Public Health. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41935438/

[11] Russell JN, Kos D, Yacoub E et al. Enhanced metagenomic surveillance for bovine respiratory disease pathogens and antimicrobial resistance by hybridization capture sequencing. Appl Environ Microbiol. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40928217/

[12] Mazzei M, Sorvillo B, Sgorbini M et al. The Role of Viral Pathogens in Horse Respiratory Diseases: A Cytological and Molecular Approach Using Next-Generation Sequencing. Animals (Basel). 2024. URL: https://pubmed.ncbi.nlm.nih.gov/39682313/

[13] Chae JB, Kim WG, Song S et al. Molecular characterization of ten viral pathogens causing calf diarrhea in Hanwoo (Bos Taurus coreanae) by next generation sequencing. BMC Vet Res. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40660160/

[14] Cao L, Wang X, Zhou Y et al. Diagnosis of Paralytic Rabies by Metagenomics Next-Generation Sequencing: A Case Report and Review of the Literature. Vet Med Sci. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41556741/

[15] Shao D, Li Y, Chen S et al. Improved RNA-based metagenomic sequencing for rapid pathogen detection in pets. BMC Vet Res. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41526915/

[16] Ferreira R, Coelho L, Santos JD et al. Probe-based metagenomic pathogen detection: advancing laboratory capacity for complex diagnosis. Front Microbiol. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/41164002/

[17] Bergner L, Catalano S, Nichols J et al. Quantifying viral load and characterizing virus diversity in wildlife samples with target enrichment sequencing. Microb Genom. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40996787/

[18] Merritt B, Ratcliff JD, Ta S et al. TaxTriage: an open-source metagenomic sequencing data analysis pipeline enabling putative pathogen detection. Bioinformatics. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41923365/

[19] Liu L, Hakhverdyan M, Wallgren P et al. An interlaboratory proficiency test using metagenomic sequencing as a diagnostic tool for the detection of RNA viruses in swine fecal material. Microbiol Spectr. 2024. URL: https://pubmed.ncbi.nlm.nih.gov/39162509/ *** Disclaimer: This article is for educational and informational purposes only. It is not intended to substitute for professional veterinary advice, diagnosis, treatment, or regulatory guidance. Always consult a licensed veterinarian or qualified specialist regarding animal health, disease diagnosis, and therapeutic decisions.

[20] Ngo C, Morrell JM, Tummaruk P. Boar semen microbiome: Insights and potential implications. Anim Reprod Sci. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/39577267/

[21] Li J, Baumgartner W, Wang L. Histopathologic and Genomic Characterization of a Novel Caprine Astrovirus Identified in a Boer Goat Kid in Illinois, United States. Viruses. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41600881/

[22] Zamparo S, Brocca G, Marroni F et al. Metabarcoding Reveals a Potentially Undescribed Columnaris-Causing Bacterium in Peracute Skin Disease of Rainbow Trout (Oncorhynchus mykiss, Walbaum). J Fish Dis. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/40579799/

[23] Kuhar U, Krapež U, Slavec B et al. Virome analysis suggests deltacoronaviruses and picornaviruses as a probable cause of severe intestinal disease on three quail farms. Anim Microbiome. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40450375/

[24] Yang B, Zhang Y, Su S et al. Metagenomic next-generation sequencing identifies tick-carried pathogens on Tarim Red Deer in Southern Xinjiang. Open Vet J. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/41630746/

[25] Perez LJ, Perez-Restrepo LS, Ciuoderis K et al. Emergence, persistence, and positive selection of yellow fever virus in Colombia. Front Microbiol. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40260085/

[26] Assenmacher CA, Mou K, Li G et al. Actinomyces sp. detected by next-generation sequencing in paraffin-embedded, formalin-fixed tissues of a dog with severe panophthalmitis and periocular cellulitis. J Vet Diagn Invest. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41562140/

[27] Wheelahan JW, Vaz PK, Legione AR et al. Virological investigation and comparative genomic analysis of elephant endotheliotropic herpesvirus 1B infection in an Australian captive herd of Asian elephants (Elephas maximus). PLoS One. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42213733/

[28] Bilhalva LC, Yacoub MN, Dos Santos AP et al. First report of Castellaniella spp. infection in dogs and the genomic evidence of a novel species. J Appl Microbiol. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41604225/

[29] Huo S, Liu W, Lv C et al. The re-emergence of psittacosis in China: a scoping review of epidemiology, diagnostics, and One Health priorities. Sci One Health. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42255501/

[30] Hake N, von Holtum C, Höper D et al. Identification and Long-Term Detection of Hepacivirus bovis Genotype 1 and 2 on a Cattle Farm in Germany. Viruses. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/41600842/

[31] Storms S, Lim A, Savard C et al. Identification of Hunnivirus in Bovine and Caprine Samples in North America. Viruses. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/41305512/

[32] Trogu T, Carrera M, Tolini C et al. Identification of Reassortant Mammalian Orthoreovirus Strains in European Hedgehogs (Erinaceus europaeus): Genomic Insights and Host Association. Microorganisms. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/41011379/

[33] Ishibashi N, Akase Y, Watanabe S et al. Genomic and molecular epidemiological characterization of a novel piscine poxvirus in the red seabream, Pagrus major. J Virol Methods. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40712891/

[34] Podowski JC, Forrester S, Yaqub T et al. Genomic reconstruction of Bacillus anthracis from complex environmental samples enables high-throughput identification and lineage assignment in Pakistan. Microb Genom. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40549420/

[35] Nesti DR, Hayashida K, Sugi T et al. Development of a semicomprehensive detection method for paramyxoviruses and its validation using Indonesian bats. Sci Rep. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40450092/