Metagenomic Sequencing Services for Clinical-Scale Microbiome Research: Shotgun Metagenomics, Viromics, and Microbial Community Analysis

A research team running a 200-participant dietary intervention trial collects stool samples at four timepoints. They want to know not just which bacteria are there, but what those bacteria are doing — which metabolic pathways they carry, whether they harbor antibiotic resistance genes, and how the microbial ecosystem shifts in response to diet. Another group, tracking an outbreak of unexplained respiratory illness in a hospital, needs to identify every potential pathogen — bacterial, viral, and fungal — in bronchoalveolar lavage samples, without knowing in advance what they are looking for. A third team monitors groundwater microbial communities near a former industrial site, searching for shifts that signal bioremediation progress.

These three projects have one thing in common: they need more than a list of bacterial genera. They need the full genomic content of every microbe in their samples. That is what shotgun metagenomic sequencing delivers.

Unlike 16S rRNA amplicon sequencing, which reads a single marker gene to identify which bacteria are present, shotgun metagenomics fragments and sequences all the DNA in a sample — bacterial, archaeal, fungal, viral, and even host-derived. The result is not just a census of who is there, but a catalog of what the community is genetically capable of doing. CD Genomics offers Metagenomic Sequencing Services spanning the full range — from shallow shotgun for cost-sensitive cohort studies to deep metagenomics with metagenome-assembled genomes, and from DNA viromics to integrated multi-omics analysis.

This guide covers when shotgun metagenomics is the right tool, how to design a project that answers your question without wasting budget, and what to expect from sample submission through bioinformatic delivery.

Beyond 16S — What Shotgun Metagenomics Unlocks

The 16S rRNA gene has been the foundation of microbial ecology for decades. It is inexpensive, the databases are mature, and the workflow is straightforward. But 16S has hard limits. It tells you which bacteria are present, usually at the genus level. It tells you nothing about what those bacteria can do. It misses fungi, viruses, and protozoa entirely. And it is blind to the mobile genetic elements — plasmids, integrons, transposons — that spread antibiotic resistance between species.

Shotgun metagenomics removes these limits by sequencing everything. Fragmenting all the DNA in a sample and sequencing the pieces without targeting a specific gene means you capture every genome present. The payoff comes in three forms.

First, taxonomic resolution to the species and strain level. Because shotgun reads are distributed across entire genomes rather than a single gene, you can distinguish Escherichia coli from Escherichia albertii, or trace a specific Clostridioides difficile ribotype through a hospital ward. For clinical research, strain-level tracking matters — it distinguishes a harmless commensal from a toxigenic strain carrying the tcdA and tcdB genes.

Second, direct functional profiling. Instead of predicting metabolic pathways from 16S data using tools like PICRUSt2 — which infer function by looking up what close relatives are known to do — shotgun metagenomics reads the actual genes. You get the KEGG pathways present in your sample, the carbohydrate-active enzymes (CAZy), the antibiotic resistance genes (CARD, ResFinder), and the virulence factors. You can see whether a gut community is enriched for butyrate synthesis genes or whether a soil sample carries novel beta-lactamases. This is not prediction. It is measurement (1, 7).

Third, coverage across all domains of life. A single shotgun library captures bacteria, archaea, fungi, DNA viruses, and microbial eukaryotes simultaneously. For samples where the fungal or viral fraction matters — gut mycobiome studies, phage ecology, agricultural soil food webs — 16S alone leaves most of the biology on the table.

The trade-off is cost. Shotgun metagenomics costs roughly two to four times more per sample than 16S amplicon sequencing, because you need millions of reads per sample rather than tens of thousands. But the cost gap is narrowing. Shallow shotgun metagenomics, which generates about half a gigabase of data per sample — roughly three million read pairs — now provides species-level taxonomic resolution and basic functional profiling at a cost approaching that of 16S. For cohort studies where genus-level 16S data is insufficient but deep shotgun is too expensive for hundreds of samples, shallow shotgun is the emerging middle ground.

CD Genomics' 16S/18S/ITS Amplicon Sequencing service provides the targeted approach for projects where genus-level bacterial profiling meets the research need, while our shotgun metagenomics services add the functional and multi-kingdom dimension when the research question demands it.

Shotgun Metagenomics for Gut Microbiome Research

The human gut harbors the most intensively studied microbial ecosystem on Earth. Shotgun metagenomics has transformed gut microbiome research from a cataloging exercise — "what lives in the gut?" — into a mechanistic science that asks: "what are these microbes producing, and how does it affect the host?"

What Shotgun Data Adds to Gut Studies

A 16S survey of stool samples from an inflammatory bowel disease cohort will tell you that Faecalibacterium is depleted in Crohn's disease patients. That is useful. A shotgun metagenomic analysis of the same samples will tell you that the lost Faecalibacterium strains carry the genes for butyrate synthesis via the acetyl-CoA pathway, that the depletion correlates with reduced fecal butyrate measured by metabolomics, and that the remaining Faecalibacterium in the patients are a different strain than those in healthy controls — one with a mutation in a key butyrate kinase gene. That is actionable (3, 4).

This functional resolution matters for intervention studies. When a probiotic, prebiotic, or dietary intervention shifts the gut community, shotgun metagenomics captures the shift in two dimensions simultaneously: which species change in abundance, and which metabolic pathways are gained or lost. For dietary fiber supplementation trials, you can directly measure whether the genes for complex carbohydrate degradation — the CAZy families that break down arabinoxylan or resistant starch — increase in abundance. For fecal microbiota transplantation studies, you can track whether donor strains engraft and whether the functional capacity of the recipient's microbiome converges toward the donor's.

CD Genomics provides Metagenomic Sequencing Services for gut microbiome studies at scales ranging from 20-subject pilot experiments to 500-subject clinical cohorts, with standardized bioinformatic pipelines that deliver taxonomic profiles, functional annotations, and strain-level tracking from stool, biopsy, and lavage samples.

Study Design That Works

The most common mistake in gut metagenomics is under-powering. The inter-individual variation in gut microbiome composition is enormous — two healthy people can differ as much in their microbial profiles as a healthy person and someone with disease. A study with ten subjects per group is unlikely to find anything except the largest effects. Thirty to fifty subjects per group is a more realistic minimum for a case-control comparison. For longitudinal studies where each subject serves as their own control — pre- versus post-intervention — smaller numbers are feasible because between-subject variation is removed.

A second design consideration is host DNA. Stool samples are forgiving — typically less than one percent of reads map to the human genome. But biopsy samples, mucosal scrapings, and tissue can contain over ninety percent host DNA. Without host depletion before sequencing, most of your data budget goes to reading the human genome, not the microbial one. Options include differential lysis of human cells before DNA extraction, or commercial host depletion kits that selectively methylate or cleave human DNA. For low-microbial-biomass biopsy samples, CD Genomics can incorporate host DNA depletion into the standard sample preparation workflow.

For a deeper dive into gut microbiome study design and analysis, see our companion guide on Shotgun Metagenomics for Gut Microbiome Studies.

Environmental Metagenomics — Reading Earth's Microbial Operating System

A single gram of agricultural soil contains more microbial species than there are fish species in the ocean — and the vast majority have never been cultured. Environmental metagenomics is the only way to access this diversity, but it comes with challenges that clinical microbiome samples do not pose: humic acids that co-extract with DNA and inhibit sequencing enzymes, degraded DNA from environmental exposure, and a handful of abundant taxa that can dominate the sequencing output and mask rare community members. But the rewards are proportionally large: environmental metagenomes are the source of most known antibiotics, the engine of global nutrient cycles, and an early-warning system for ecosystem stress.

What Environmental Metagenomics Reveals

A gram of agricultural soil typically yields between ten and fifty million shotgun reads after quality filtering. From those reads, you can reconstruct the nitrogen cycling capacity of the community — the relative abundance of genes for nitrogen fixation, nitrification, denitrification, and anammox. You can assess whether a community carries the genetic potential to degrade specific pollutants — petroleum hydrocarbons, chlorinated solvents, pesticides. And you can discover novelty: environmental metagenomics is the primary source of new enzymes, new biosynthetic gene clusters, and new microbial phyla. In 2025 alone, metagenome-assembled genomes from environmental samples expanded the known bacterial tree of life by several candidate phyla that have never been cultured.

For environmental monitoring, shotgun metagenomics detects community shifts that 16S misses. A 16S survey might show that the relative abundance of Proteobacteria increased at a contaminated site. A metagenomic analysis of the same samples would show that the increase is driven by specific Betaproteobacteria carrying catechol dioxygenase genes for aromatic hydrocarbon degradation — confirming that the shift is a functional response to contamination, not a random fluctuation.

Handling the Challenges of Environmental Samples

Environmental DNA extraction is harder than stool or saliva extraction. Soil, sediment, and water filters contain PCR inhibitors — humic acids, heavy metals, complex polysaccharides — that co-purify with DNA. Specialized extraction kits that include inhibitor removal steps are essential. For low-biomass samples — deep subsurface groundwater, oligotrophic open ocean water, polar ice cores — the risk of contamination from laboratory reagents and the environment swamping the true biological signal is high. Negative controls — extraction blanks and library preparation blanks sequenced alongside the real samples — are not optional. They are the only way to distinguish a real microbial signal from reagent contamination.

CD Genomics offers Metagenomic Sequencing Services tailored for environmental samples, with validated extraction protocols for soil, sediment, freshwater, marine, and extreme environment matrices, including inhibitor removal and negative control workflows.

Beyond pollution monitoring, environmental metagenomics is driving discovery. The majority of new antibiotics developed in the past decade trace their origins to biosynthetic gene clusters first identified in soil metagenomes. Metagenome mining — computationally scanning assembled contigs for clusters of genes that look like they could produce secondary metabolites — has become a systematic alternative to traditional culture-based natural product discovery. The approach scales to thousands of samples and does not require the organisms to be culturable. A single gram of soil can yield dozens of novel biosynthetic gene clusters, any one of which could encode a new antimicrobial or anticancer compound (5).

For a comprehensive treatment of environmental metagenomics from sampling strategy to functional annotation, see our guide on Environmental Metagenomics — Soil, Water, and Sediment.

Viromics — Sequencing the Viral Fraction

Viruses are the most abundant biological entities on Earth. In the ocean, there are roughly ten viruses for every bacterial cell. In the human gut, the virome — the community of viruses, dominated by bacteriophages — is estimated to contain more unique genes than the bacterial microbiome. And in clinical samples, identifying a viral pathogen when you do not know which virus to look for is the hardest diagnostic problem in infectious disease research.

DNA and RNA Viromes

Shotgun metagenomics of total DNA captures DNA viruses — bacteriophages, herpesviruses, adenoviruses, papillomaviruses — alongside bacterial and fungal genomes. But many clinically and ecologically important viruses have RNA genomes: influenza, SARS-CoV-2, norovirus, rotavirus, and the vast diversity of RNA phages. Capturing the RNA virome requires a separate library preparation step that reverse-transcribes RNA into complementary DNA before sequencing. A complete virome analysis therefore often involves two libraries: a DNA library for DNA viruses and phages, and an RNA library for RNA viruses.

Bacteriophages dominate most viromes. In the human gut, the phage community is highly individual-specific — your gut phage profile is more unique to you than your bacterial profile — and remarkably stable over time (6). Phages influence gut bacterial community dynamics through predation, and they mediate horizontal gene transfer by moving DNA between bacterial hosts. When a gut community is disrupted by antibiotics, phage populations often bloom, and the phage-encoded antibiotic resistance genes they carry can spread through the recovering bacterial community (6, 9).

Virus Detection Without Target Primers

The key advantage of metagenomic viromics over targeted PCR panels is that you do not need to guess which viruses to look for. A respiratory panel tests for twenty or thirty known pathogens. Shotgun metagenomics sequences everything — known and unknown — and identifies viruses by matching reads against viral genome databases. For outbreak investigations where the causative agent is unknown, for research studies of encephalitis or sepsis where culture-based and PCR-based pathogen screening has been exhausted, and for environmental surveillance where novel viruses emerge from animal reservoirs, this untargeted approach is the only way to capture the full picture.

The challenge is sensitivity. Viral nucleic acids can account for less than 0.01% of total DNA or RNA in a clinical sample. Enrichment strategies — filtering samples to remove bacterial and host cells by size, depleting host DNA, or using capture probes that target conserved viral sequences — can increase the viral fraction by orders of magnitude. CD Genomics incorporates virus-enrichment protocols into its virome sequencing workflow for samples where viral load is expected to be low.

For an in-depth treatment of virome sequencing methods and applications, see our guide on Viral Metagenomics and Virome Sequencing.

Phage therapy research is another area where viromics is gaining traction. As multidrug-resistant bacterial infections outpace the development of new antibiotics, phages — viruses that infect and kill specific bacteria — offer an alternative that does not select for antibiotic resistance. Identifying therapeutic phages starts with screening environmental or clinical viromes for phages that target the pathogen of interest. Metagenomic viromics is the discovery engine that powers this pipeline.

Planning a Metagenomics Project — Practical Decisions

The decisions that determine whether a metagenomics project succeeds are made before the first sample enters the sequencer. The most important ones are straightforward.

Sequencing Depth: How Much Data Per Sample?

Sequencing depth is measured in gigabase pairs per sample — the total amount of DNA sequenced. For a typical human stool sample, here is what each depth tier delivers.

Shallow shotgun metagenomics, at roughly 0.5 to 1 gigabase per sample, provides species-level taxonomic profiles and detects the most abundant functional genes. It is the cost-effective choice for large cohort studies where the primary goal is to identify compositional differences between groups.

Standard shotgun metagenomics, at 5 to 10 gigabases per sample, adds comprehensive functional profiling — you see the full complement of KEGG pathways, CAZy families, and antibiotic resistance genes. It also provides enough coverage to assemble metagenome-assembled genomes — draft genomes of individual microbial species reconstructed computationally from the mixed community sequencing data.

Deep metagenomics, at 20 gigabases or more per sample, is reserved for projects that require near-complete genomes from rare community members, comprehensive virome characterization, or discovery of novel biosynthetic gene clusters. Deep sequencing of a few representative samples often complements shallow sequencing of the full cohort.

The right depth is the minimum that answers your question. Sequencing deeper than necessary adds cost without adding insight. Our team can help you estimate the required depth based on your sample type and research goals.

Host DNA — The Silent Budget Killer

For samples from human tissue, saliva, skin swabs, or biopsies, host DNA can consume the majority of your sequencing reads. If ninety percent of your reads map to the human genome, you effectively paid for ten times less microbial data than you thought. Host depletion — removing human DNA before library preparation — is strongly recommended for any sample where host cells are expected to outnumber microbial cells. Options include differential centrifugation, selective lysis of human cells, and commercial kits that use methylation-specific nucleases to degrade human DNA while leaving microbial DNA intact.

Bioinformatics — What You Get and What You Need

After sequencing, the raw data passes through a bioinformatic pipeline that removes low-quality reads, filters out any remaining host DNA, and then assigns taxonomic identities and functional annotations to the remaining reads. The standard deliverables include an interactive taxonomic report showing the community composition at phylum through species level, a functional profile mapping genes to KEGG pathways and other databases, and diversity analyses that quantify differences between sample groups.

For projects that require custom analysis — strain-level tracking, metagenome-assembled genome reconstruction, or integration with metabolomics or metatranscriptomics data — CD Genomics provides tailored bioinformatic support. The most common pitfall is underestimating the computational resources needed for large metagenomics projects. Assembling metagenomes from fifty samples at standard depth requires substantial memory and processing power, and trying to run these analyses on a laptop is a recipe for frustration. Cloud-based analysis or access to high-performance computing is the practical standard.

Cost and Service Options

Once the experimental design is settled — sample numbers, sequencing depth, and analysis scope — the next question is practical: what does this cost, and what service model fits your lab's capabilities?

Metagenomic sequencing costs are dominated by three factors: the number of samples, the sequencing depth per sample, and the level of bioinformatic analysis required. Understanding how these levers interact makes the difference between a project that fits the budget and one that does not.

Sequencing-Only vs. Full-Service

Sequencing-only means you provide extracted DNA, and the service provider handles library preparation, sequencing, and delivery of raw data files. You handle all downstream analysis yourself or through a separate bioinformatics service. This is the most cost-effective route for labs with in-house bioinformatic expertise and computational infrastructure.

Full-service includes DNA extraction from your samples, library preparation, sequencing, and the complete bioinformatic analysis — from quality control through taxonomic and functional annotation to publication-ready figures. Full-service costs more per sample but eliminates the need for specialized personnel and computing hardware. For clinical research groups that generate microbiome data but do not employ dedicated bioinformaticians, full-service is typically the more efficient route.

Between these two ends of the spectrum, CD Genomics offers modular service packages. You can send us extracted DNA but request full bioinformatic analysis. You can have us perform host depletion on your tissue samples while you handle the data analysis. The goal is to match the service level to the capabilities already present in your lab, so you pay only for what you need.

Budgeting for a Metagenomics Project

For a typical human stool metagenomics study, here is a rough cost framework. At the shallow shotgun level — around one gigabase per sample — you can expect to pay roughly one hundred to two hundred dollars per sample for library preparation and sequencing, with bioinformatic analysis adding proportionally. Standard-depth shotgun metagenomics runs roughly two to three times that for sequencing, reflecting the additional data generated. These are ballpark figures for budgeting purposes. Actual pricing depends on sample numbers, sample type, required depth, and analysis scope.

To put these numbers in context: a 50-sample shallow shotgun study of a dietary intervention cohort, with standard bioinformatic analysis — taxonomic profiling plus KEGG functional annotation — typically falls in a total budget range that is competitive with deep 16S amplicon sequencing at similar cohort sizes, while delivering species-level resolution and direct functional gene detection that 16S cannot provide. For budget planning, the key variables are sample count — the dominant multiplier — sequencing depth per sample, and analysis scope.

The most effective way to control costs is to invest in good study design up front. Five common mistakes inflate budgets: over-sequencing samples beyond what the question requires, under-sequencing a pilot and having to re-do the full cohort, forgetting to budget for data storage and computational costs, skipping host depletion and wasting reads on human DNA, and failing to include negative controls that catch contamination before it ruins a sequencing run.

CD Genomics' project consultation team works with you during the experimental design phase to align sequencing strategy with budget. For cohort studies where shallow shotgun metagenomics provides sufficient information for the primary analysis, we can design a staged approach: shallow sequencing on the full cohort, followed by deep metagenomics on a subset of samples selected based on the initial results. For projects that combine community-level metagenomics with isolate-level whole genome sequencing — for example, metagenomic screening followed by whole genome sequencing of cultured isolates of interest — CD Genomics' Whole Genome Sequencing services provide the complementary isolate-level view.

For broader context on how metagenomics fits into the microbial genomics landscape, see our Amplicon Sequencing Services Hub, which covers 16S, 18S, ITS, and DNA barcoding approaches. For projects that combine metagenomics with isolated bacterial genomes, our Bacterial Whole Genome Sequencing Guide covers de novo assembly, re-sequencing, and mutation detection from pure cultures.

FAQ

When should I choose shotgun metagenomics over 16S amplicon sequencing?

Choose shotgun metagenomics when you need species- or strain-level identification, when you need to know the functional capacity of the community — what genes and pathways are present — or when you need to capture fungi, viruses, and other non-bacterial microbes alongside bacteria. Choose 16S when your question is limited to "which bacteria are present, and how do their abundances differ between groups," and budget is the primary constraint.

What is shallow shotgun metagenomics, and how does it compare to 16S?

Shallow shotgun metagenomics generates less data per sample — typically 0.5 to 1 gigabase — which is enough for species-level taxonomic profiling and detection of the most abundant functional genes. It costs more than 16S but less than standard shotgun, and it provides species-level resolution that 16S does not, plus direct detection of a subset of functional genes. For large cohort studies where species-level taxonomy matters but deep functional profiling is not required, it is the emerging cost-effective compromise.

Can shotgun metagenomics replace culturing?

No. Metagenomics tells you what genes are present in a sample, but it does not tell you which microbes are viable, which are metabolically active, or how they behave in isolation. Metagenomics and culturing are complementary. A metagenomics survey can tell you which organisms to target for isolation, and culturing can validate functional predictions made from metagenomic data — for example, confirming that a Bacteroides strain carrying a predicted polysaccharide utilization locus actually grows on that polysaccharide.

How much does host DNA matter in clinical samples?

It matters enormously. Tissue biopsies, skin swabs, and saliva samples can contain over ninety percent human DNA. Without host depletion, most sequencing reads map to the human genome rather than the microbiome, dramatically reducing the effective microbial sequencing depth. For samples where host DNA is expected to exceed fifty percent, host depletion should be part of the standard workflow.

How many samples do I need for a metagenomics study?

It depends on the effect size you are looking for and the inter-individual variability in your population. For case-control studies of the human gut microbiome, thirty to fifty subjects per group is a practical minimum. For longitudinal studies where each subject serves as their own control, smaller numbers are feasible. For environmental surveys where the goal is to characterize a site rather than compare groups, even a few well-chosen samples can provide valuable data — but biological replication is always better than technical replication.

What kind of bioinformatic support will I need?

If you choose sequencing-only service, you will need a bioinformatician or computationally skilled researcher who can run quality control, taxonomic classification, and functional annotation pipelines. Experience with the Linux command line, a basic understanding of scripting, and access to a server or cloud computing resources are the minimum. If you choose full-service, CD Genomics handles the entire computational workflow and delivers analyzed results — no command line required.

What is the turnaround time for a typical metagenomics project?

For a standard project of fifty to one hundred samples, expect four to six weeks from sample receipt to analyzed data delivery for library preparation and sequencing, plus an additional two to four weeks for full bioinformatic analysis. Expedited timelines are available for time-sensitive projects. Large cohort studies with hundreds of samples require proportionally longer sequencing time but benefit from batch processing that keeps per-sample turnaround manageable.

Can I integrate metagenomics with other omics data?

Yes. Metagenomics is frequently combined with metabolomics to link microbial gene content to metabolite profiles, or with metatranscriptomics to distinguish actively expressed genes from those merely present in the community. CD Genomics offers multi-omics integration services that correlate metagenomic functional profiles with metabolomic or metatranscriptomic data from the same samples, providing a systems-level view of microbial community function.

References:

1. Zhang L, Chen F, Zeng Z, et al. Advances in metagenomics and its application in environmental microorganisms. Frontiers in Microbiology. 2021;12:766364. doi:10.3389/fmicb.2021.766364
2. Blanco-Miguez A, Beghini F, Cumbo F, et al. Extending and improving metagenomic taxonomic profiling with MetaPhlAn 4. Nature Biotechnology. 2023;41(11):1633-1644. doi:10.1038/s41587-023-01688-w
3. Almeida A, Nayfach S, Boland M, et al. A unified catalog of 204,938 reference genomes from the human gut microbiome. Nature Biotechnology. 2021;39(1):105-114. doi:10.1038/s41587-020-0603-3
4. Kim N, Ma J, Kim J, et al. Genome-resolved metagenomics: a game changer for microbiome medicine. Experimental & Molecular Medicine. 2024;56:1501-1512. doi:10.1038/s12276-024-01262-7
5. Nayfach S, Roux S, Seshadri R, et al. A genomic catalog of Earth's microbiomes. Nature Biotechnology. 2021;39(4):499-509. doi:10.1038/s41587-020-0718-6
6. Gregory AC, Zablocki O, Zayed AA, et al. The Gut Virome Database reveals age-dependent patterns of virome diversity in the human gut. Cell Host & Microbe. 2020;28(5):724-740.e8. doi:10.1016/j.chom.2020.08.003
7. Meyer F, Fritz A, Deng ZL, et al. Critical Assessment of Metagenome Interpretation: the second round of challenges. Nature Methods. 2022;19(4):429-440. doi:10.1038/s41592-022-01431-4
8. Gu W, Miller S, Chiu CY. Clinical metagenomic next-generation sequencing for pathogen detection. Annual Review of Pathology. 2019;14:319-338. doi:10.1146/annurev-pathmechdis-012418-012751
9. Camarillo-Guerrero LF, Almeida A, Rangel-Pineros G, Finn RD, Lawley TD. Massive expansion of human gut bacteriophage diversity. Cell. 2021;184(4):1098-1109.e9. doi:10.1016/j.cell.2021.01.029
10. Jimenez-Arroyo C, Molinero N, Sabater C, Margolles A, Delgado S, Ruas-Madiedo P. Human gut microbiome study through metagenomics: recent advances and challenges for clinical implementation. Enfermedades Infecciosas y Microbiologia Clinica (English ed.). 2025;43(6):322-331. doi:10.1016/j.eimce.2025.03.007

For research purposes only, not intended for clinical diagnosis, treatment, or individual health assessments.