Figure 1: The animal microbiome spans diverse livestock systems — from the rumen of cattle to the gut of monogastric species and aquatic environments — each requiring tailored metagenomic strategies for meaningful biological interpretation.

A livestock nutritionist formulating a feed additive trial, a veterinary researcher tracking antimicrobial resistance genes across a poultry flock, and an aquaculture geneticist screening for growth-promoting gut bacteria in shrimp all share the same underlying need: they need to know which microbes are present, what they are doing, and whether shifts in community composition correlate with the phenotype they care about. Metagenomic sequencing answers these questions, but the path from sample to actionable result depends on picking the right method for the right question. This guide walks through the three main sequencing strategies — 16S rRNA amplicon, ITS amplicon, and shotgun metagenomics — as they apply to cattle, swine, poultry, and aquaculture research, with practical guidance on sample collection, sequencing depth, and bioinformatics deliverables that hold up under peer review.

Why the Animal Microbiome Matters for Livestock Production

The microbial communities inhabiting the gastrointestinal tract of livestock animals influence feed conversion efficiency, immune development, pathogen colonization resistance, and even behavior. In ruminants, the rumen microbiome ferments otherwise indigestible plant fiber into volatile fatty acids that supply up to 70% of the animal's energy. In monogastric species — pigs and poultry — the hindgut microbiota shape nutrient absorption, gut barrier integrity, and systemic inflammation. In aquaculture, the gut microbiome of farmed fish and shrimp modulates growth rate, disease susceptibility, and water quality tolerance.

Researchers who quantify these microbial communities gain a molecular-level view of processes that manifest at the whole-animal level as weight gain, illness, or mortality. A metagenomic dataset from a feed trial, for example, can reveal whether a candidate probiotic actually engrafts in the gut or whether a dietary intervention shifts the ratio of fiber-degrading to starch-fermenting bacteria in the rumen. Without sequencing, these questions remain black-box observations of phenotype with no mechanistic substrate.

Studies that pair microbiome data with host genomic or transcriptomic data add an additional layer of resolution. The crosstalk between the rumen epithelium and its resident microbiota, reviewed comprehensively by Ye and colleagues, illustrates how host gene expression and microbial metabolite profiles jointly regulate feed efficiency and methane output in cattle [1]. For research teams working across multiple omics layers, the epigenomic sequencing resource for agricultural research covers chromatin-level regulation that may interact with microbial signals in the gut.

16S rRNA vs. ITS vs. Shotgun Metagenomics: Which Approach Fits Your Question

Choosing a sequencing strategy starts with a clear definition of the research question. Each method answers a different type of question at a different resolution, and the cost per sample can vary by an order of magnitude across platforms.

A practical workflow for many livestock microbiome projects uses a tiered approach: screen 50 to 200 fecal or rumen samples with 16S rRNA sequencing at 50,000 reads per sample to identify treatment-associated shifts in community composition, then select 10 to 20 samples representing the extremes of the response for shotgun metagenomic sequencing at 20 million reads per sample. This design captures the statistical power of amplicon-based community profiling while reserving the higher cost of shotgun sequencing for the samples where functional annotation adds the most biological value.

For teams needing fungal community profiling alongside bacterial data — for example, a study examining how ration changes alter both bacterial fermentation and fungal fiber degradation in the rumen — pairing 16S and ITS amplicon sequencing from the same DNA extract provides complementary taxonomic information. CD Genomics offers both 16S/ITS rRNA amplicon sequencing and shotgun metagenomic sequencing as standalone or combined services.

Sample Collection That Preserves Microbial Community Composition

Sample handling introduces more variation into microbiome datasets than any other pre-sequencing variable. A fecal sample left at room temperature for four hours will produce a community profile that reflects post-collection growth of facultative anaerobes, not the in vivo community. Standardizing collection, preservation, and storage across all samples and time points is non-negotiable.

Commercial stabilization kits (DNA/RNA Shield, OMNIgene·GUT, and RNAlater) are acceptable alternatives to snap-freezing when liquid nitrogen is not available in the field. However, each preservative introduces its own extraction bias — switching preservatives mid-study invalidates cross-timepoint comparisons. Pick one preservation method at the start and use it for every sample.

Metadata fields that should accompany every sample include: animal ID, species, breed, age, sex, diet or ration group, treatment group, sampling date and time, sample type, preservation method, and storage duration before extraction. Missing metadata is the most common reason livestock microbiome datasets cannot be re-analyzed or compared across studies.

Rumen Microbiome Profiling for Feed Efficiency and Methane Reduction

The rumen is the most intensively studied microbial ecosystem in livestock science. A single milliliter of rumen fluid contains roughly 10¹⁰ bacteria, 10⁶ archaea, 10⁵ protozoa, and 10³ fungi, organized into a metabolic network that degrades plant structural polysaccharides into short-chain fatty acids. The composition of this network directly affects how efficiently a steer converts feed into body mass and how much methane the animal produces per kilogram of dry matter intake.

Shotgun metagenomic studies of the rumen have identified specific bacterial genera — Prevotella, Fibrobacter, Ruminococcus, and Butyrivibrio — whose relative abundance correlates with residual feed intake, a heritable measure of feed efficiency. Multi-breed host rumen epithelium transcriptome and microbiome analyses by Fonseca and colleagues demonstrated that differentially expressed microbial genes involved in hydrogen metabolism and methanogenesis distinguish high-efficiency from low-efficiency animals, providing functional targets for microbiome-informed breeding or dietary intervention [2]. Projects that need to resolve strain-level differences in these key taxa should incorporate shotgun sequencing rather than relying solely on 16S amplicon data.

For breeding programs and nutrition trials that collect rumen samples from dozens or hundreds of animals, 16S rRNA sequencing at the V3–V4 region provides sufficient taxonomic resolution for beta-diversity analyses (Bray-Curtis, weighted UniFrac) that test whether diet, breed, or treatment group explains microbial community variation. When the research question extends to which microbial genes — not just which microbial genera — drive differences in feed efficiency or methane output, shotgun metagenomics at 20–40 million reads per sample becomes necessary to capture the functional gene repertoire.

Teams combining rumen microbiome profiling with host transcriptomic data from rumen epithelium may find the outsourcing transcriptome analysis guide for crop and livestock research useful for planning the RNA-seq component of a multi-omics study design.

Gut Microbiome and Host Health in Monogastric Livestock

In pigs and poultry, the gut microbiome occupies the distal small intestine, cecum, and colon, where it ferments residual dietary fiber into short-chain fatty acids, synthesizes B vitamins and vitamin K, and competitively excludes enteric pathogens. Unlike ruminants, monogastric animals have a single-compartment stomach where low pH kills most ingested microbes before they reach the small intestine — meaning the hindgut community is shaped primarily by diet composition, host genetics, and housing environment, not by incoming feed-associated microbes.

Tang and colleagues characterized the fecal metagenome of commercial pigs across multiple production stages, identifying functional shifts in carbohydrate-active enzyme (CAZyme) gene families that track age-related changes in diet formulation [3]. Their data showed that weaning — the transition from milk to solid feed — triggers a reproducible restructuring of the gut microbiome that can be monitored by 16S rRNA sequencing at 50,000 reads per sample. Shotgun metagenomics on a subset of these samples further revealed that the post-weaning gut metagenome is enriched for genes involved in starch and simple sugar metabolism, consistent with the cereal-based composition of grower-finisher diets.

In broiler chickens, Peña and colleagues applied shotgun metagenomics to cecal content samples and found that birds with higher body weight carried distinct assemblages of Lactobacillus species and showed enrichment for microbial genes in amino acid biosynthesis pathways, suggesting that the cecal microbiome contributes to host protein nutrition beyond what feed formulation alone predicts [4]. For poultry researchers, cecal content collected at necropsy provides a more stable and representative microbial community profile than fecal droppings collected from litter, where environmental contamination and aerobic degradation can obscure treatment effects.

A frequent question in monogastric microbiome studies is whether to target the luminal or mucosa-associated fraction. The answer depends on the biology: luminal contents capture the community passing through the gut and reflect diet-driven changes, while mucosa-associated communities capture the bacteria that physically interact with the host epithelium and are more likely to influence immune function and barrier integrity. Collecting both fractions from the same gut segment doubles the sample count but provides complementary information that a single fraction cannot.

Aquaculture Microbiome Research: From Gut to Pond Water

The gut microbiome of farmed fish and shrimp functions in an aquatic environment where the boundary between host-associated and environmental microbes is more permeable than in terrestrial livestock. Water constantly flushes the intestinal lumen, and the gill surface area rivals that of the gut epithelium as a microbiome interface. This permeability means that aquaculture microbiome studies must sample not only gut contents but also rearing water and, in the case of filter-feeding species like shrimp, pond sediment.

Kanika and colleagues reviewed the expanding literature on fish gut metagenomics, documenting consistent patterns in which herbivorous and omnivorous species harbor more diverse and carbohydrate-fermenting gut communities than carnivorous species, whose gut metagenomes are enriched for proteolytic functions [5]. Their analysis highlighted that 16S rRNA sequencing remains the workhorse for aquaculture microbiome surveillance, but shotgun metagenomics is gaining traction for functional questions — such as identifying microbial genes involved in chitin degradation in shrimp or characterizing antimicrobial resistance gene reservoirs in fish farm effluent.

For researchers designing aquaculture microbiome experiments, the key logistical consideration is sample stability between collection and freezing. On a commercial fish farm or shrimp pond, the distance between the sampling site and a −80°C freezer can be measured in hours. DNA/RNA Shield or RNAlater stored at ambient temperature for up to 7 days is validated for both 16S and shotgun metagenomic workflows and removes the need to carry liquid nitrogen into the field. However, these preservatives add cost per sample that should be budgeted at the study design stage.

Pooling individual fish or shrimp into composite samples — for example, gut contents from 5 fish per tank, homogenized into a single extraction — is a common strategy for reducing sequencing costs while capturing tank-level biological variability. The trade-off is that individual-level variation within a tank is lost. When the research question concerns individual animal performance (for example, correlating individual growth rate with gut microbiome composition), samples should be extracted and sequenced individually rather than pooled.

Figure 3: Three major livestock microbiome research domains — rumen fermentation in cattle, gut health in monogastric species, and aquaculture microbiome dynamics — each supported by distinct metagenomic sequencing strategies and sample collection protocols.

From Raw Reads to Biological Insight: Bioinformatics Considerations

The bioinformatics pipeline for amplicon and shotgun metagenomic data differs substantially, and the choice of pipeline should be made before samples are submitted, not after data are in hand.

For 16S rRNA and ITS amplicon data, the standard workflow uses QIIME2 or mothur for quality filtering, denoising (DADA2 or Deblur), amplicon sequence variant (ASV) calling, and taxonomic assignment against reference databases such as SILVA (16S) or UNITE (ITS). Downstream analyses include alpha diversity (Shannon, Chao1, observed ASVs), beta diversity (Bray-Curtis, weighted and unweighted UniFrac), and differential abundance testing (ANCOM-BC, DESeq2, or LEfSe). Amplicon data can also be used to predict functional potential through tools like PICRUSt2, which imputes KEGG pathway abundances from 16S taxonomic profiles — a useful screening tool that should be interpreted as hypothesis-generating, not confirmatory.

For shotgun metagenomic data, the pipeline branches into taxonomic profiling (Kraken2, MetaPhlAn4) and functional annotation (HUMAnN 3, eggNOG-mapper). Taxonomic profiling of shotgun data provides species- and strain-level resolution that 16S data cannot match, while functional annotation directly quantifies gene family and pathway abundances without the inference step required for amplicon data. The downstream output includes KEGG ortholog abundance tables, CAZyme family profiles, and antimicrobial resistance gene counts mapped against databases such as CARD or ResFinder.

For livestock microbiome projects that require both amplicon and shotgun data, bioinformatics integration can be streamlined by using a single agricultural genomic data analysis service that handles both data types, ensuring consistent quality filtering, taxonomic assignment, and statistical modeling across the full dataset. Researchers who lack in-house bioinformatics capacity or who want peer-review-ready figures and statistical reports can also work with a dedicated bioinformatics analysis service to go from raw FASTQ files to publication-quality visualizations.

Designing a Livestock Microbiome Study That Holds Up Under Review

A common pattern in rejected microbiome manuscripts is insufficient attention to experimental design before the first sample is collected. The most frequent criticisms from reviewers — underpowered group sizes, unaccounted batch effects, missing negative controls — are all preventable if the study design is locked down at the protocol stage.

Replicate numbers cited above reflect community-level effect sizes (Bray-Curtis R² of 0.05–0.10 between treatment groups) observed in published livestock microbiome studies. Studies targeting smaller effect sizes or interactions between diet, breed, and treatment should increase group sizes accordingly. A pilot study with 5–6 samples per group and 16S rRNA sequencing can inform power calculations before committing to a full-scale shotgun metagenomic experiment.

Batch effects — where samples collected, extracted, or sequenced on different dates cluster by batch rather than by treatment — are a serious threat to livestock microbiome studies because farm visits often happen in waves, and it is rarely feasible to collect all samples in a single day. The best defense is to randomize treatment and control group sample collection across farm visits rather than collecting all treatment samples on one visit and all control samples on another. If complete randomization is logistically impossible, equal numbers of treatment and control animals should be sampled on each visit.

For studies that include shotgun metagenomics, host DNA contamination should be assessed before library preparation. Fecal and rumen samples from cattle and pigs routinely yield 50–90% host DNA, which reduces the effective microbial sequencing depth and wastes sequencing budget on host reads. Host DNA depletion kits or deeper sequencing to compensate for host read loss should be planned during the budgeting phase. The animal-and-plant de novo genome assembly guide discusses sequencing strategies relevant to host genome reference construction, which can inform host read filtering in metagenomic analyses.

Workflow Overview: What Happens After You Submit Your Samples

Figure 2: The metagenomic sequencing workflow — from sample receipt through DNA extraction, library construction, sequencing, and bioinformatics analysis to final report delivery.

Step 1 — Sample intake and quality check. Samples arrive frozen on dry ice or in preservation buffer at ambient temperature. Each sample is logged, assigned a unique identifier, and inspected for tube integrity, labeling clarity, and sufficient volume. Samples that arrive thawed, under-filled, or unlabeled are flagged and the submitter is contacted before proceeding.

Step 2 — DNA extraction. Extraction method is matched to sample type. Fecal and rumen samples are extracted with bead-beating protocols that lyse Gram-positive bacteria and fungi; mucosa samples may require enzymatic pre-treatment to release epithelium-associated microbes. A DNA extraction blank (extraction reagents processed without a sample) is included in every batch to track kit contamination.

Step 3 — Library preparation. For 16S/ITS amplicon sequencing, the target region (V3–V4 for bacteria, ITS1 or ITS2 for fungi) is amplified with barcoded primers, pooled, and purified. For shotgun metagenomics, extracted DNA is fragmented, end-repaired, adapter-ligated, and amplified. Host DNA depletion may be performed at this stage if the sample type warrants it.

Step 4 — Sequencing. Amplicon libraries are sequenced on the Illumina MiSeq or NovaSeq platform to the agreed read depth. Shotgun libraries are sequenced on the NovaSeq to produce 2×150 bp paired-end reads at the contracted read count. A PhiX spike-in control lane is included to monitor run performance.

Step 5 — Bioinformatics analysis. Raw reads are demultiplexed, quality-filtered, and processed through the pipeline selected at project initiation (QIIME2 for amplicon; Kraken2 + HUMAnN 3 for shotgun). The output includes taxonomic abundance tables, diversity metrics, differential abundance testing, and functional annotation tables.

Step 6 — Report delivery. The final data package is delivered via secure download link and includes raw FASTQ files, processed abundance tables, diversity analysis figures, statistical test outputs, a methods section draft suitable for manuscript insertion, and a QC summary report with instrument-run metadata.

Frequently Asked Questions About Livestock Microbiome Sequencing

How many biological replicates do I need for a livestock microbiome study?

For group comparison studies using 16S rRNA sequencing, eight to ten animals per treatment group is the minimum to detect moderate community-level differences. If the expected effect size is small or if the study includes a diet-by-breed interaction term, plan for twelve to fifteen animals per group. For shotgun metagenomics, five to eight animals per group can be sufficient for functional gene-level comparisons because gene abundance data have lower inter-individual variability than taxonomic abundance data. The most reliable approach is to run a pilot 16S rRNA study with the same animal population and use the observed within-group variability to power the full experiment.

Can I combine 16S rRNA and ITS sequencing in the same project?

Yes. Both 16S rRNA and ITS amplicon libraries can be prepared from the same DNA extract using separate primer sets and then pooled on the same sequencing run with distinct barcodes. A single DNA extraction from a fecal or rumen sample yields enough material for both bacterial and fungal profiling. The sequencing depth per library should remain in the 30,000 to 50,000 read range for each amplicon type.

What sequencing depth is recommended for 16S rRNA versus shotgun metagenomics in livestock samples?

For 16S rRNA sequencing of fecal or rumen samples, 50,000 reads per sample is sufficient to capture the dominant and subdominant taxa in most livestock gut communities. For samples with expected low microbial biomass, such as small-intestine mucosa, increasing to 100,000 reads per sample improves detection of rare taxa. For shotgun metagenomics, 20 million reads per sample is adequate for species-level taxonomic profiling and broad functional annotation. Increasing to 40 to 60 million reads per sample is recommended when the research question requires strain-level resolution, antimicrobial resistance gene detection, or assembly of metagenome-assembled genomes.

How do I handle host DNA contamination in fecal or gut content samples?

Fecal and gut content samples from cattle and pigs routinely contain 50 to 90 percent host DNA by mass. This host DNA competes with microbial DNA during library preparation and sequencing, reducing the effective microbial read count. Three strategies are available. First, use a host DNA depletion kit during library preparation to selectively remove host DNA while retaining microbial DNA. Second, perform deeper sequencing to compensate for host read loss — calculate the required total reads by dividing the target microbial read count by the estimated microbial DNA fraction. Third, apply computational host read filtering during bioinformatics by mapping all reads against the host reference genome and discarding aligning reads. The first approach is the most cost-effective for projects with high expected host contamination.

What bioinformatics deliverables should I expect from a microbiome sequencing project?

A complete bioinformatics deliverable for an amplicon sequencing project includes raw FASTQ files, an ASV or OTU count table, taxonomic assignment tables at phylum through genus level, alpha and beta diversity metrics with statistical test results, principal coordinates analysis plots, differential abundance testing results, and a methods section draft. For a shotgun metagenomics project, the deliverable additionally includes functional annotation tables with KEGG ortholog and pathway abundances, CAZyme family profiles, and antimicrobial resistance gene detection results if requested. All processed data files are provided in CSV or TSV format, and figures are delivered as publication-ready PDFs or PNGs.

How CD Genomics Supports Livestock Microbiome Projects

A livestock microbiome project runs on logistics as much as science — coordinating sample shipments from multiple farm sites, tracking metadata across dozens or hundreds of animals, and producing a data package that is clean enough to analyze and transparent enough to publish. CD Genomics provides end-to-end support for animal microbiome studies, from study design consultation and sample collection kit distribution through sequencing and bioinformatics analysis.

The service portfolio covers all three sequencing strategies discussed in this guide. For bacterial community profiling, the 16S/ITS rRNA sequencing service supports high-throughput cohort studies at 30,000 to 100,000 reads per sample with QIIME2-based bioinformatics. For functional metagenomic analysis, the metagenomics sequencing service provides shotgun sequencing at 20 to 60 million reads per sample with Kraken2 taxonomic profiling and HUMAnN 3 functional annotation. For projects that require custom analytical approaches — such as metagenome-assembled genome binning, strain-level tracking across time points, or multi-omics integration — the agricultural genomic data analysis service offers tailored bioinformatics pipelines. Researchers who prefer to handle the bioinformatics in-house but need library preparation and sequencing can also submit samples for sequencing only and receive demultiplexed FASTQ files as the deliverable.

A complementary resource for researchers working on soil-associated microbiome questions — for example, comparing rhizosphere and bulk soil communities in forage crop systems — is the page on soil metagenomics for sustainable agriculture, which covers 16S and shotgun strategies for soil rather than animal samples.

For Research Use Only. CD Genomics agricultural microbiome sequencing services are provided for research purposes and are not intended for clinical diagnostic or therapeutic use. Results support scientific investigation and publication; they do not constitute veterinary medical advice or treatment recommendations.

References

1. Ye X, Sahana G, Lund M, Cai Z, et al. "The crosstalk between host and rumen microbiome in cattle: insights from multi-omics approaches and genome-wide association studies." World J Microbiol Biotechnol, vol. 41, 2025, Article 145. doi:10.1007/s11274-025-04504-6
2. Fonseca P, Lam S, Chen Y, Waters S, Guan LL, Cánovas A, et al. "Multi-breed host rumen epithelium transcriptome and microbiome associations and their relationship with beef cattle feed efficiency." Sci Rep, vol. 13, 2023, Article 20458. doi:10.1038/s41598-023-43097-8
3. Tang Q, Yin X, Wen G, Luo Z, Zhang L, Tan S, et al. "Unraveling the composition and function of pig gut microbiome from metagenomics." Anim Microbiome, vol. 7, 2025, Article 19. doi:10.1186/s42523-025-00419-7
4. Peña N, Lafuente I, Sevillano E, Feito J, Allendez G, Muñoz-Atienza E, Crispie F, Cintas L, et al. "Exploring the functional potential of the broiler gut microbiome using shotgun metagenomics." Genes, vol. 16, no. 8, 2025, Article 946. doi:10.3390/genes16080946
5. Kanika NH, Liaqat N, Chen H, Ke J, Lu G, Wang J, Wang C, et al. "Fish gut microbiome and its application in aquaculture and biological conservation." Front Microbiol, vol. 16, 2025, Article 1521048. doi:10.3389/fmicb.2024.1521048