If you have been told an infection needs "sequencing" to identify it, or you are a clinician weighing molecular testing options, the terminology can be confusing — NGS, 16S, shotgun, metagenomics, mNGS. This guide explains what next-generation sequencing is, walks through how 16S rRNA sequencing and clinical metagenomics each work from the laboratory bench to the bioinformatics pipeline, compares what each can and cannot detect, and lays out the current regulatory and insurance landscape for this kind of testing.
What Is Next-Generation Sequencing (NGS)?
Next-generation sequencing (NGS) is a technology that reads millions of fragments of DNA or RNA in parallel, allowing scientists to determine the genetic "sequences" present in a sample quickly and at scale. In infectious disease, NGS lets a laboratory identify microbes by their genetic material rather than by growing them in culture.
Older sequencing methods read one DNA fragment at a time. Next-generation sequencing (NGS) — sometimes called high-throughput or massively parallel sequencing — reads enormous numbers of fragments simultaneously, which is what makes it practical for use in healthcare. For diagnosing infections, this type of testing is powerful because it does not depend on culturing the organism: microbes that grow slowly, do not grow in standard culture, or have been suppressed by prior antibiotics can still be detected by their nucleic acids, the building blocks of DNA and RNA. This enables detection of microbial DNA or RNA directly from clinical specimens such as urine, blood, and cerebrospinal fluid.
NGS is the shared foundation beneath both methods this page compares. The critical difference is what portion of the genetic material is sequenced. That single decision separates 16S rRNA sequencing from clinical metagenomics and drives everything downstream.
16S rRNA Sequencing: Bacterial Detection
16S rRNA sequencing identifies bacteria by sequencing one specific gene — the 16S ribosomal RNA gene — that all bacteria share but that varies enough between species to act as an identifier. It is a targeted, amplicon-based method, meaning it amplifies and reads only that one gene region.
The 16S rRNA gene is present in essentially all bacteria and archaea, and it contains regions that are highly conserved (nearly identical across species) alternating with "hypervariable" regions that differ between organisms. By reading the variable regions, a lab can assign the bacteria present to taxonomic groups. Because it targets a universal bacterial gene, 16S is a long-established, relatively economical way to survey the bacterial makeup of a sample.
From the Laboratory Bench: How 16S Works
The laboratory workflow is targeted and amplification-driven:
- Sample collection and DNA extraction: Total DNA is extracted from a biological specimen.
- PCR amplification: Universal PCR primers (sequences of DNA that "prime" the region to create copies) bind the conserved regions flanking one or more hypervariable regions (commonly labeled V1–V9). PCR then amplifies the target regions, making millions of copies.
- Library preparation: The amplicons, or copies of the target region, are prepared for sequencing, which occurs in a separate laboratory machine called a sequencer.
- Sequencing: The amplicons are sequenced on an NGS platform, which outputs a large amount of data containing the "sequence" of the nucleic acids for the targeted region.
While useful in some cases, particularly in resource-limited settings or for specific research goals, 16S rRNA sequencing is an inherently limited approach for diagnosing infections. Because the test is only designed to amplify the 16S rRNA gene, it can only detect bacteria and archaea. Moreover, because there is a wealth of other information included in bacterial genomes that is not captured through this method, including antimicrobial resistance markers and virulence factors, the ability of 16S sequencing to provide strain-level insights is next to none. This is important because knowing the specific strain may change the treatment strategy for certain bacteria.
Bioinformatics: How 16S Works
Once sequencing is done, the data generated from the sequencing platform is run through a bioinformatic pipeline, or a series of steps that use different methods to appropriately clean, format, and interpret the data. The bioinformatic pipeline typically runs: quality filtering and trimming of low-quality reads; denoising or clustering of sequences into amplicon sequence variants (ASVs) or operational taxonomic units (OTUs); and taxonomic assignment by comparing those sequences to a curated 16S reference database, which may be public or proprietary. The usual output is a taxonomic profile, most reliably to the genus level; species-level resolution is often not achievable from a single short 16S region.
Taxonomy prediction, or predicting which bacteria are present, depends heavily on the reference database used. Coverage of bacteria in 16S databases is currently more mature than whole-genome databases, however, more coverage does not always translate to better in clinical contexts. Many databases used for 16S sequencing contain bacterial species rarely, if ever, found in humans. Unless carefully accounted for, this can cause issues of misclassification, whereby closely related species are misidentified. This can also increase the risk of false-positives via contamination introduced during sample collection and laboratory processing. Not only do these issues bring into question the validity of the results, they also make interpretation by healthcare providers more difficult by expanding the total list of bacteria present — even those that may not be clinically relevant.
What 16S rRNA Sequencing Can and Cannot Do
Its strengths are cost, simplicity, maturity, and well-curated bacterial reference databases. Its constraints are consequential for clinical pathogen identification: it detects bacteria and archaea only (no fungi, viruses, or parasites); resolution is usually limited to genus rather than species or strain; it does not directly reveal antimicrobial resistance or virulence genes; and it carries primer bias — organisms whose target region does not match the "universal" primers well can be under-detected or missed. Because 16S reads only the 16S rRNA gene and suffers incomplete primer coverage, clinical metagenomic sequencing approaches achieve greater cross-domain coverage.
Oftentimes, 16S rRNA sequencing is combined with separate PCR testing to identify antimicrobial resistance genes or to expand coverage to non-bacterial microorganisms. This can only partially address its limited approach, as PCR testing is similarly a targeted strategy. If the PCR test was not designed to look for a specific microbe or an important gene, then it will not find it.
Clinical Metagenomics: Comprehensive Infectious Disease Detection and Characterization
Clinical metagenomics — also called shotgun metagenomic sequencing — sequences all DNA present in a sample at once, rather than a single target gene found across bacteria and archaea. This lets it detect not only bacteria and archaea, but viruses, fungi, and parasites as well.
Unlike 16S, clinical metagenomics allows for detailed species and strain-level insights; not only for identification of specific microorganisms, but also for antimicrobial resistance marker and virulence factor detection. It also sequences the human DNA present in the sample, though the usefulness of this information for diagnosing and treating infections has yet to be determined.
Because clinical metagenomic-based tests make no assumption about what to look for in a biological sample, it is described as hypothesis-free or untargeted: it can detect an unexpected or novel pathogen that a targeted test was never designed to find. This enables simultaneous, hypothesis-free detection of a broad array of pathogens directly from clinical specimens, unlike 16S rRNA sequencing and culture-based testing.
From the Laboratory Bench: How Clinical Metagenomics Works
For clinical metagenomic sequencing, the laboratory workflow is untargeted:
- Sample collection and nucleic acid extraction: Total DNA (and in some instances, RNA, to capture RNA viruses) is extracted from the biological sample.
- Host depletion / enrichment (often): Clinical samples are dominated by human DNA, which can make it difficult to read signals from the microorganisms present. To counter this, many workflows include a step to reduce host material and/or enrich for microbial DNA. By doing these steps, the proportion of human to microbial DNA in the sample shifts towards containing a greater proportion of microbial DNA. This helps generate more useful data downstream.
- Random fragmentation and library preparation: Following host depletion/enrichment, all of the DNA is fragmented. Library preparation follows, which is a multi-step process that converts the fragments into a specific format to be read by the sequencer.
- Deep sequencing: The "library" of DNA fragments is then sequenced, typically far more deeply than 16S, to capture low-abundance organisms against the host background. This produces the data that is then put through a bioinformatics pipeline, similar to 16S.
The absence of a specific target is exactly what gives clinical metagenomics its breadth — and what makes host DNA, cost, and sequencing depth its key practical hurdles in the laboratory.
Bioinformatics: How Clinical Metagenomics Works
The bioinformatics pipelines used in clinical metagenomic sequencing are substantially heavier than 16S. A typical pipeline includes: quality control and trimming; removal of human/host reads (important for both accuracy and patient-privacy reasons); taxonomic classification of the remaining microbial reads against comprehensive genome databases (could be public and/or proprietary); and, frequently, screening for antimicrobial resistance markers and virulence factors. This more substantial bioinformatics pipeline provides species/strain-level identification plus functional gene content.
Clinical metagenomics has a different set of limitations as compared to 16S. These heavier pipelines often take more computing resources or longer to complete, unless specifically optimized for. Additionally, though clinical metagenomics is often called "unbiased," its workflows are still subject to biases introduced during sample preparation, library construction, and bioinformatic analysis, all of which can affect sensitivity and taxonomic resolution. This is why it is important to understand whether a clinical metagenomic-based diagnostic test has been rigorously analytically and clinically validated.
What Clinical Metagenomics Can and Cannot Do
Its strengths define the clinical case for it: cross-domain detection (bacteria, viruses, fungi, parasites in one test), species and often strain-level resolution, detection of resistance and virulence genes, and the ability to find unexpected or hard-to-culture organisms without a prior hypothesis. The value of clinical metagenomic-based diagnostics is demonstrated in hard-to-diagnose infections, such as infections that return culture-negative, involve multiple pathogens (polymicrobial infections), or for patients at higher risk of complications from an infection, such as those with a compromised immune system.
Its constraints are the flip side of that power: higher cost per test; heavy interference from host DNA (especially in low-biomass or high-host samples, such as cerebrospinal fluid or blood); and greater bioinformatic complexity. Similar to 16S, distinguishing a true pathogen from harmless colonizers, contaminants, or environmental background is also an issue — though this is often mitigated in clinical metagenomic workflows through bioinformatic optimizations, careful clinical curation of pathogen databases, and validation of thresholds to call a microbe present.
The Key Differences: 16S rRNA Sequencing vs. Clinical Metagenomics
The core difference: 16S sequences the 16S rRNA gene only, while clinical metagenomics sequences all genetic material in a sample. As a result, clinical metagenomics detects a far broader range of organisms (including viruses and fungi), resolves them to species or strain level, and can provide insight into antimicrobial resistance and virulence factors. 16S remains cheaper and simpler, but clinical metagenomic sequencing provides the more complete clinical picture.
Both methods use next-generation sequencing; the difference is scope. 16S sequencing detects only part of the microbial community that shotgun sequencing reveals, and with sufficient sequencing depth, shotgun sequencing has more power to identify less-abundant organisms. This is why, for clinical pathogen identification where breadth and precision matter, clinical metagenomics is generally the more comprehensive tool.
That said: 16S retains genuine advantages in cost, workflow simplicity for the laboratory, and — for some applications — more mature reference databases. The two are not strictly "better vs. worse"; they answer different questions. For the untargeted, "what is causing this infection when we do not know where to look" problem, clinical metagenomics is the stronger fit.
| Feature | 16S rRNA sequencing | Clinical metagenomics (shotgun / mNGS) |
|---|---|---|
| What's sequenced | One marker gene (16S rRNA) | All DNA (sometimes RNA) in the sample |
| Organisms detected | Bacteria and archaea only | Bacteria, archaea, fungi, viruses, parasites |
| Typical resolution | Genus level, sometimes species | Species / strain level |
| Resistance genes | Not detected | Comprehensively detected |
| Approach | Targeted (amplicon) | Untargeted (hypothesis-free) |
| Primer bias | Yes | No target-primer bias |
| Host DNA interference | Lower | Higher (needs depletion) |
| Bioinformatic complexity | Lower | Higher |
| Relative cost | Lower | Higher |
| Reference databases | Mature for bacteria, but may cause challenges if clinically-irrelevant bacteria are included | Clinically-curated to include microbes found in the human body |
A Leader in Clinical Metagenomics: Biotia's BIOTIA-ID and BIOTIA-DX
Biotia develops clinical metagenomic sequencing-based diagnostic tests and bioinformatic pipelines to analyze sequencing data. Biotia's diagnostic platform, BIOTIA-ID, consists of the laboratory workflow used on biological samples prepared for sequencing. Biotia's bioinformatics tool, referred to as BIOTIA-DX, is the software that analyzes the sequencing data to support pathogen identification and characterization — including antimicrobial resistance markers and virulence factors.
Both have gone through rigorous research and development to optimize the laboratory and bioinformatics workflows to overcome challenges facing all clinical metagenomic-based tests. These include specialized approaches to host depletion and microbial enrichment, fine-tuning of certain bioinformatic steps to reduce the impact of contamination and set validated thresholds for positivity, and a unique machine learning step to more accurately predict specific strains, antimicrobial resistance markers, and virulence factors. Moreover, the reference database used by Biotia was curated by expert clinicians and clinical microbiologists to aid healthcare providers in interpretation of the findings.
Currently, BIOTIA-ID and BIOTIA-DX have been approved for use as a laboratory-developed test on urine specimens only. Called the BIOTIA-ID Urine Test, it was designed for patients managing unexplained or unresolved urinary tract infection symptoms. Patients who meet the clinical criteria for recurrent UTI, complicated UTI, or culture-negative UTI may purchase an at-home collection kit from our website. This allows them to connect with a UTI-specialist provider from Clinova Solutions for any necessary care.
Aside from urine specimens, Biotia is currently developing a clinical metagenomic-based diagnostic test for prosthetic joint infections, using synovial fluid samples. As part of a partnership with Hospital for Special Surgery, this diagnostic test will provide patients managing unexplained or unresolved joint infections with comprehensive pathogen insights — including antimicrobial resistance profiling.
The Current Landscape: Regulation and Insurance
Clinical metagenomic tests in the U.S. are typically offered as laboratory-developed tests (LDTs), regulated under the Clinical Laboratory Improvement Amendments (CLIA, overseen by Centers for Medicare and Medicaid Services) and often accredited by the College of American Pathologists (CAP), rather than through FDA medical device clearance. Insurance coverage remains limited and inconsistent. The technology's clinical value is increasingly recognized, but reimbursement pathways lag behind.
Regulatory Status
Most clinical metagenomic diagnostic tests reach patients as laboratory-developed tests (LDTs) — designed, validated, and performed within a single lab. Recently, the U.S. Food and Drug Administration (FDA) rescinded a Final Rule that would have required laboratories to receive FDA medical device approval for LDTs. While this would have ensured the quality and safety of LDTs, it would have simultaneously created an enormous burden on diagnostic laboratories to run clinical trials for their test offerings. Because clinical trials are expensive, many specialized LDTs that patients and providers rely on were at risk of being pulled from the market entirely. However, just because an LDT is not approved by the FDA does not necessarily mean it is ineffective.
LDTs are overseen through CLIA (administered by CMS), commonly with CAP accreditation and, in some states, additional programs like New York State's Clinical Laboratory Evaluation Program. Clinical laboratories offering LDTs still must comply with CLIA requirements, and regulatory frameworks including CLIA and CAP are beginning to accommodate NGS-based diagnostic tests. Despite this regulation, the bar for quality varies state-by-state. So how do you know whether an LDT offered by a diagnostic laboratory performs at the level it says it does? There are a few ways to tell.
- The laboratory is credentialed: All clinical diagnostic laboratories are required to be CLIA-certified according to the type of testing they perform. Check to see if the laboratory's CLIA number is displayed on their website, or alternatively, look up the laboratory's CLIA number using CMS's lookup tool.
- The test is approved in New York State: Among the diagnostics industry, it is widely held opinion that New York State's Clinical Laboratory Evaluation Program's requirements are the most strict in the country — seconded only to FDA medical device clearance. If a test is able to be performed in New York State, it is likely that the evidence supporting the LDT in question is particularly strong.
- There is published evidence: Many diagnostic laboratories will publish analytical validation, clinical validation, and clinical utility studies of their LDTs in peer-reviewed, scientific journals. It is not necessary to deeply understand these publications — simply their presence can add confidence that the test performs as the laboratory says it does.
- There is clinician oversight: The FDA notes that legitimate medical testing requires "meaningful involvement" by a licensed healthcare provider who can explain the test and contextualize the results. Many direct-to-consumer tests offered by laboratories do not involve a healthcare provider. Opt for those that do.
- There is no supplement component: Many direct-to-consumer laboratory testing companies will simultaneously offer dietary supplements, with the test results guiding patients to a specific product or formulation. Many of these supplements have not been rigorously studied to understand their benefits and risks. Always look for peer-reviewed scientific research supporting the use of the specific supplement and any ingredients in its formulation for the health conditions in question.
Insurance and Reimbursement
Reimbursement is the larger real-world barrier. Unlike other diagnostic testing methods, which have well-established billing codes and reimbursement pathways, many NGS-based tests lack standardized billing mechanisms. Established CPT codes favor PCR-based single-pathogen testing, reimbursing roughly $50–150 per test, while NGS reimbursement remains highly inconsistent. Some proprietary (PLA) codes exist, but such tests are variably priced and often denied by insurers. Some commercial payers explicitly exclude it — at least one major insurer's coverage policy states plainly that metagenomic NGS is not covered or reimbursable. There are early, narrow bright spots: California's Medi-Cal program approved limited reimbursement for mNGS in suspected encephalitis or meningitis, following evidence that it improves outcomes — an important early precedent for the field. Moreover, Biotia has made progress in obtaining insurance coverage for the BIOTIA-ID Urine Test, receiving a proprietary laboratory analysis (PLA) code and moving through obtaining Medicare coverage.
The honest summary for patients and providers: NGS-based testing's clinical utility is increasingly documented, but coverage is still limited, payer-specific, and often requires prior authorization or appeal. Currently, more often than not, patients face out-of-pocket costs. Anyone considering this testing should confirm coverage with their specific plan.
