Research method
Metagenomic Sequencing
Metagenomic sequencing reads DNA from a mixed community without first culturing each species, producing gene catalogues, strain-level gene-content calls, or — when the library is actually 16S amplicons — a taxonomic census. Shotgun metagenomes can report accessory genes within a named species; 16S surveys cannot. Presence of a gene or taxon is not a host phenotype.
Microbiome studies use shotgun metagenomes when they need functions or within-species gene content, not only who is there. It answers 'which microbial genes or strains differ across people, diets, or disease states?' The main limitation is that association is not causation, sample sizes can be small, and at least one paper in this list is a 16S biogeography survey rather than shotgun metagenomics.
Evidence
What the evidence shows
Drawn from 4 studies in this library. Each finding starts with a plain-language takeaway, then the denser detail. Supports means evidence for a finding; Challenges means evidence against a stated position; Qualifies marks scope with a short note on each study’s contribution. Challenged positions are labeled — they are not findings.
A gene-deletion detection method on human gut metagenomes found that strain-level gene content for the same species varies substantially among individuals in natural habitats. Presence/absence is not a full phenotype, and sample sizes remain limited.
Stool metagenomes plus epithelial mRNA from 6 breast-fed and 6 formula-fed infants showed diet-altered Firmicutes/Actinobacteria composition. The study tested 459 intestinal and 660 immunity genes, with many differentially expressed between diets. Small n and correlation do not prove which microbes cause each transcript change.
In a longitudinal DIPP cohort, 947 stool samples from 76 HLA-risk children (29 seroconverters versus 47 controls) showed Bacteroides dorei dominance in cases prior to persistent autoantibodies (mean diagnosis age 16.8 months). That is an observational association, not proof that B. dorei causes type 1 diabetes.
Healthy-body biogeography analysed more than 24 million 16S reads from 2,983 specimens across HMP and related cohorts: 30 phyla to 929 genera, strong habitat differences, average V3–V5 depth about 8,167 reads. This is amplicon biogeography, not shotgun gene content, and it does not define disease causality for any taxon.
Open questions
Tensions and limits
Some items are genuine disagreements on the same question. Others mark different assays, populations, or outcomes — limits on how far one study travels — not a forced fight between papers.
Shotgun gene-content metagenomics (within-species deletions across people; infant diet genes plus 459/660 host transcripts; B. dorei before seroconversion) is not the same experiment as HMP-style 16S biogeography (>24 million reads, 2,983 specimens, ~8,167 V3–V5 reads). A 16S habitat map cannot report accessory genes; a gene-deletion call cannot replace body-wide taxonomy.
- How much do gut bacterial gene contents differ?
- Infant diet shapes gut–host immunity
- B. dorei rises before T1D autoimmunity
- Human body microbiome biogeography
Study Role Design N Population Outcome How much do gut bacterial gene contents differ? Supports Cross-sectionalMetagenomic gene-deletion detection quantifying within-species gene content across gut communities N=207 · 252 fecal metagenomes from 207 individuals (HMP and European MetaHIT) Human gut metagenomes from public cohorts Inter-individual within-species gene-content variation Infant diet shapes gut–host immunity Supports Cross-sectionalInfant stool metagenomes correlated with host epithelial transcriptomes by feeding mode N=12 · Six breast-fed and six formula-fed infants Breast-fed and formula-fed human infants Diet-linked microbiota composition and host intestinal/immunity gene expression B. dorei rises before T1D autoimmunity Supports CohortLongitudinal DIPP stool sequencing in HLA-risk children before T1D autoimmunity N=76 · 947 stool samples; 29 seroconverters vs 47 autoantibody-negative controls HLA-DQB1 moderate-to-high-risk children in DIPP Gut microbiota (B. dorei dominance) prior to persistent T1D autoantibodies Human body microbiome biogeography Supports Cross-sectionalMulti-habitat 16S analysis of HMP and related healthy cohorts N=236 · 2,983 specimens from 236 HMP subjects (plus smaller urine/penis, preterm stool, conjunctiva sets); >24 million reads Healthy Human Microbiome Project subjects across body habitats Taxonomic biogeography of the healthy human microbiome Even among shotgun studies, claims differ in strength. Infant diet work is n=6 versus 6 with host-transcript correlations; DIPP is 76 children and 947 samples but still does not prove B. dorei causes autoimmunity; within-species gene content varies without assigning a disease. Pooling them as 'the microbiome causes immunity' overreads every design.
- Infant diet shapes gut–host immunity
- B. dorei rises before T1D autoimmunity
- How much do gut bacterial gene contents differ?
Study Role Design N Population Outcome Infant diet shapes gut–host immunity Supports Cross-sectionalInfant stool metagenomes correlated with host epithelial transcriptomes by feeding mode N=12 · Six breast-fed and six formula-fed infants Breast-fed and formula-fed human infants Diet-linked microbiota composition and host intestinal/immunity gene expression B. dorei rises before T1D autoimmunity Supports CohortLongitudinal DIPP stool sequencing in HLA-risk children before T1D autoimmunity N=76 · 947 stool samples; 29 seroconverters vs 47 autoantibody-negative controls HLA-DQB1 moderate-to-high-risk children in DIPP Gut microbiota (B. dorei dominance) prior to persistent T1D autoantibodies How much do gut bacterial gene contents differ? Supports Cross-sectionalMetagenomic gene-deletion detection quantifying within-species gene content across gut communities N=207 · 252 fecal metagenomes from 207 individuals (HMP and European MetaHIT) Human gut metagenomes from public cohorts Inter-individual within-species gene-content variation
Common misconceptions
If B. dorei dominates before autoantibodies, that species has been shown to cause type 1 diabetes.
947 samples from 76 HLA-risk children showed dominance prior to persistent autoantibodies (mean diagnosis age 16.8 months). The finding is observational association, not causation.
16S biogeography of 929 genera is a metagenomic gene catalogue of the healthy body.
The dataset is >24 million 16S reads from 2,983 specimens (mean V3–V5 depth ~8,167). Amplicon taxonomy is not shotgun gene content and does not assign functions or disease causality.
Infant diet changed 459 intestinal and 660 immunity genes, so those microbes have been proven to drive each transcript.
n=6 breast-fed versus 6 formula-fed; taxa and genes were correlated with host mRNA. Association with small n does not identify which microbes cause each change.
Exam-style questions
Short-answer questions that ask you to explain or compare, not recall.
Why can two people carry the 'same' gut species yet differ in metagenomic gene content, and what can presence/absence still not prove?
Strain-level accessory genes vary among individuals in natural habitats, detected as metagenomic gene deletions. Presence/absence is not a full phenotype and sample sizes remain limited.
DIPP found B. dorei dominance before seroconversion at a mean diagnosis age of 16.8 months in HLA-risk children. What extra evidence would be needed to claim causation?
Experimental colonisation or a trial that changes B. dorei and then autoimmunity, plus exclusion of confounders. 29 seroconverters versus 47 controls in 947 samples is still observational.
A student cites '30 phyla to 929 genera' as shotgun metagenomics of disease. What was actually sequenced, and what claim is barred?
>24 million 16S reads from 2,983 healthy-cohort specimens, habitats differing strongly, ~8,167 V3–V5 reads on average. Disease causality for any taxon is not defined.
Infant metagenomes changed Firmicutes/Actinobacteria with diet while hundreds of host immunity genes differed. How do you keep the host and microbe layers from being collapsed into one causal arrow?
Diet altered both community composition and host epithelial mRNA (459 intestinal and 660 immunity genes tested) in 6 versus 6 infants. Correlation across layers does not prove which microbes caused which transcripts.
The studies
4 studies in this library bear on Metagenomic Sequencing, ordered by citations.
- Human body microbiome biogeography
Healthy human body sites host distinct bacterial communities across a massive 16S survey.
- B. dorei rises before T1D autoimmunity
Bacteroides dorei dominates the gut microbiome before autoimmunity in high-risk Finnish children.
- Infant diet shapes gut–host immunity
Breast- vs formula-fed infants differ in gut microbiota and host immune gene expression.
- How much do gut bacterial gene contents differ?
Same gut bacterial species can differ widely in gene content across people, shaping functional individuality.
Learn alongside