Prevention & care
Microbiome tests: sequencing, relative abundance, reproducibility and clinical utility
On this page
Microbiome profiling uses a specimen, often stool, to characterize microorganisms or their genetic material. A report may describe the community, compare it with a reference dataset or propose health advice. Those are separate steps: measuring microbial DNA does not automatically validate a diagnosis or a personalized treatment.[1]
Name the method and the reported quantity
| Test or quantity | What it examines | Interpretation boundary |
|---|---|---|
| 16S amplicon sequencing | Sequences a selected microbial marker region | The assay and analysis workflow affect the profile[2] |
| Whole metagenome shotgun sequencing | Sequences a wider collection of microbial genetic material | This is a different workflow from targeted marker sequencing[2] |
| Relative abundance | A component’s proportion of the reported composition | A changing percentage need not mean its absolute amount increased[4] |
| Quantitative profiling | Adds information about microbial load | It asks a different question from proportions alone[5] |
A sequence profile is the output of a chain of methods
The Microbiome Quality Control project studied blinded stool, artificial-community and other specimens across 15 laboratories and nine analysis protocols. Variation depended on specimen type and origin, DNA extraction, sample-handling environment and computational analysis. A reported organism list therefore reflects preparation and classification choices as well as the original specimen. This does not make sequencing meaningless. It explains why method documentation, suitable controls and reproducibility are necessary before reports from different workflows are treated as interchangeable.[3]
A percentage is not a count of all organisms
Gloor and colleagues explain that sequencing datasets are compositional. Reported components share a constrained total, so one component’s proportion can increase when another decreases, even without growth of the first component. Comparisons require methods that account for this structure. A colorful chart of relative abundance should therefore not be read as a direct census of absolute organisms in the entire intestine. The denominator and the material sampled matter as much as the prominent percentage.[4]
Absolute load changes the biological question
Vandeputte and colleagues combined microbial profiling with cell counting to study how community composition relates to microbial load. They found that relative-only profiling can obscure relationships when total load differs between samples. This research supports distinguishing ratios from quantities. It does not establish a universal target count for healthy consumers or show that a larger number of organisms always implies better health. Ask whether the test actually measured load or only reported proportions.[5]
Evidence & uncertainty
Analytical reliability is a prerequisite for clinical interpretation. A February 2026 study examined seven consumer microbiome services using standardized fecal material and found substantial differences within and between providers. The study tests comparability; it does not establish that every present-day service has the same performance.[2]
Selected-source overview · Independent review pendingPublished questions
Does a microbiome health score establish a medical diagnosis?
Selected-source overview · Independent review pending. Read the evidence context.
No universal interpretation is established. The consensus cautions against strict healthy species-abundance ranges and routine clinical use without supporting evidence. Ask how the score was validated for the intended decision and population.[1]
Evidence behind this answer
- International consensus on microbiome testing in clinical practice
clinical guideline · 2025
Original indexed consensus abstract and selected statement passages inspected; PMC open returned a browser challenge; no independent full-text appraisal · General principles; healthy ranges; clinical utility; therapeutic-advice boundary
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Does a bigger bacterial percentage prove more of that bacterium grew?
Selected-source overview · Independent review pending. Read the evidence context.
No. Relative abundance is compositional. A proportion can rise because other components fall. Absolute quantity requires additional information; a percentage alone cannot settle that question.[4]
Evidence behind this answer
- Microbiome datasets are compositional
methods or conceptual explanation · 2017
Original indexed methodological explanation inspected; PMC open returned a browser challenge; no independent full-text appraisal · Compositional structure, ratios and analysis implications
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Should identical stool material give identical consumer profiles?
Selected-source overview · Independent review pending. Read the evidence context.
It should produce sufficiently reproducible measurements for the intended use. The 2026 comparison found important differences across and within seven services. This is a reliability finding about those assessed workflows, rather than a verdict on every future service.[2]
Evidence behind this answer
- Analytical performance of consumer gut-microbiome services
laboratory or methods study · 2026
Selected original full-text methods and results inspected; no independent full-text appraisal · Abstract; standardized sample, three kits per provider, workflows and reproducibility
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Does relative profiling always measure the total microbial load?
Selected-source overview · Independent review pending. Read the evidence context.
No. Quantitative profiling adds load information that proportions alone lack. Research shows this distinction can change biological interpretation. A report should identify which quantity was measured rather than imply that relative abundance directly provides absolute counts.[5]
Evidence behind this answer
- Quantitative profiling links gut-community variation to load
laboratory or methods study · 2017
Original abstract and selected explanatory passages inspected; no independent full-text appraisal · Abstract: cell counts, relative profiling and microbial load
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Why can sample handling and analysis change an organism report?
Selected-source overview · Independent review pending. Read the evidence context.
Extraction, handling and computational classification contribute variation. The quality-control study examined these effects across laboratories. Comparing profiles requires attention to the workflow and controls, not just the organism names in two charts.[3]
Evidence behind this answer
- Microbiome Quality Control project: variation in amplicon sequencing
laboratory or methods study · 2017
Original indexed abstract and selected methodological findings inspected; publisher full text unavailable; no independent full-text appraisal · Original abstract: blinded specimens, 15 laboratories, nine pipelines and sources of variation
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
Is finding a microbe the same as showing it causes symptoms?
Selected-source overview · Independent review pending. Read the evidence context.
No. CDC notes that C. difficile molecular tests can be positive in people without infection. A relevant clinical indication and appropriate interpretation distinguish organism detection from a diagnosis of disease.[6]
Evidence behind this answer
- Clinical testing and diagnosis for C. difficile infection
agency condition or prevention explanation · 2024
Selected relevant original official educational passages inspected · Molecular detection, asymptomatic positives and pre-test probability
This answer summarizes selected sources. An independent appraisal of the complete evidence has not been completed.
What the consumer-service comparison tested
The 2026 study sent three kits per service using a homogenized pool of human fecal material. Companies were not informed before processing and reporting. Differences between providers were on a scale comparable with differences between donors in the authors’ comparison. These findings raise a concrete reliability concern. However, a pooled-material assessment at one time is not a clinical trial of test-guided treatment, a comprehensive audit of every provider or proof about all future methods.[2]
A targeted pathogen test has a different indication
CDC describes indication-specific C. difficile tests for toxin-related genes, antigen or toxins. Molecular testing can detect an organism in people without disease, including asymptomatic carriers. Symptoms and pre-test probability are part of interpretation. This illustrates why a clinically selected infection test differs from a broad wellness community profile: they have different targets and intended decisions. A microbe’s appearance on a broad report should not independently prompt an antibiotic or an antimicrobial supplement.[6]
A plausible biological association is not a validated treatment pathway
The 2025 international consensus states that evidence is insufficient to widely recommend routine microbiome testing in clinical practice. It also cautions that strict healthy reference ranges for species proportions are generally unsupported and discourages providers from attaching therapeutic advice to reports. The panel describes a framework for developing useful tests, rather than declaring that microbiome science has no future. Important gaps include whether profiling improves an actual clinical decision and whether acting on it improves outcomes compared with suitable usual care.[1]
References
- International consensus on microbiome testing in clinical practice
Clinical guideline
International multidisciplinary microbiome-testing consensus panel · Content/publication date: 2025
Accessed: 2026-10-13 UTC · Original indexed consensus abstract and selected statement passages inspected; PMC open returned a browser challenge; no independent full-text appraisal
Cited sections: General principles; healthy ranges; clinical utility; therapeutic-advice boundaryExpert Delphi consensus, not a treatment-effect trial or an atlas-assigned evidence grade
- Analytical performance of consumer gut-microbiome services
Laboratory or methods study
Servetas and colleagues · Content/publication date: 2026
Accessed: 2026-10-13 UTC · Selected original full-text methods and results inspected; no independent full-text appraisal
Cited sections: Abstract; standardized sample, three kits per provider, workflows and reproducibilityPublished February 26, 2026; an earlier NIST listing describes prepublication work
- Microbiome Quality Control project: variation in amplicon sequencing
Laboratory or methods study
Sinha and colleagues · Content/publication date: 2017
Accessed: 2026-10-13 UTC · Original indexed abstract and selected methodological findings inspected; publisher full text unavailable; no independent full-text appraisal
Cited sections: Original abstract: blinded specimens, 15 laboratories, nine pipelines and sources of variation - Microbiome datasets are compositional
Methods or conceptual explanation
Gloor and colleagues · Content/publication date: 2017
Accessed: 2026-10-13 UTC · Original indexed methodological explanation inspected; PMC open returned a browser challenge; no independent full-text appraisal
Cited sections: Compositional structure, ratios and analysis implications - Quantitative profiling links gut-community variation to load
Laboratory or methods study
Vandeputte and colleagues · Content/publication date: 2017
Accessed: 2026-10-13 UTC · Original abstract and selected explanatory passages inspected; no independent full-text appraisal
Cited sections: Abstract: cell counts, relative profiling and microbial load - Clinical testing and diagnosis for C. difficile infection
Agency condition or prevention explanation
CDC · Content/publication date: 2024
Accessed: 2026-10-13 UTC · Selected relevant original official educational passages inspected
Cited sections: Molecular detection, asymptomatic positives and pre-test probability
Scope & review
- Population
- General readers learning about clinical testing and consumer test claims; no individualized diagnosis
- Exposure or intervention
- Microbiome tests
- Comparison
- Indication-specific alternatives; screening, diagnostic evaluation and monitoring distinguished
- Outcomes
- Meaning of measurements, appropriate indications, collection requirements and limits of inference
- Timeframe
- A specimen or examination at a stated time; trends and follow-up depend on clinical context
- Jurisdiction
- Selected US agency sources, dated professional guidance and original international research; jurisdiction and eligibility are source-specific
Selected-source educational explanation. Official guidance, original study findings and editorial interpretation are distinguished. No exhaustive search, independent systematic synthesis or qualified medical review has been completed.
Sources checked on their individual access dates. Selected sources were checked; ongoing surveillance and qualified review have separate statuses on the coverage page.
Revision history
- 2026-10-13 UTC: Added a selectively sourced diagnostic-test explanation; qualified independent review remains pending.
What this article still leaves open
- Which profiling methods add independently replicated clinical benefit for a defined indication?
- How reproducible are current reports across collection, storage, analysis and repeat sampling?
Suggest a correction
Submit a content correction to the editorial queue, or download a local note. Leave out personal medical details. Submitted text is stored privately with a browser receipt session and is visible to this browser and the editor. Clearing cookies removes access to your submission history. This is not an account or a medical service.
Your submitted corrections for this article
Explore the topic map
Browse the health domains connected to this article. The map also identifies topics whose articles remain unwritten.