Concept
Microbiome
15 studies
The microbiome is the community of microorganisms living in or on a host (or in an environment), studied for composition, genes, and effects on host physiology. In this library it ranges from human gut taxa linked to metabolic traits and disease risk, through strain-level gene content and bile-salt enzymes, to experimental animal and bee systems that test how disruptions change the community.
Students meet microbiome claims that sound interchangeable—“gut bugs cause disease”—when the evidence is usually a specific host, assay, and outcome. The skill is saying what a study adds, where it applies, and what would change the position.
Studies
15
Findings
9
9 supporting · 0 challenging · 17 qualifying citations
Open tensions
3
Currently
What we know
- Genetic instruments for gut taxa link a small set of microbiome–cancer associations under Mendelian randomisation.
- Before type 1 diabetes autoimmunity, Bacteroides dorei can dominate children’s gut communities.
- In middle-aged men, gut composition associates with metabolites and metabolic-syndrome traits.
- Bacterial bile-salt enzymes can change host bile acids and weight-related phenotypes in models.
- Strain-level gene content varies a lot within the same gut species — species labels are coarse.
Largest unresolved question
Cancer MR, pediatric T1D timing, and adult metabolic-trait associations all involve gut microbiota, but they answer different disease and design questions. Treating them as one interchangeable “gut microbiome disease effect” collapses distinct estimands.
Common misconceptions
Mendelian randomisation proving a taxa–cancer association means the microbe’s mechanism in tumours is known.
MR supports a genetic-instrument association under its assumptions; the cancer MR paper notes unproven experimental pathways and limited strain resolution from 16S-based taxa.
B. dorei dominance before autoantibodies proves that B. dorei causes type 1 diabetes.
The finding is a temporal observational association before autoimmunity; it does not by itself establish causation.
If two papers both study “the gut microbiome,” their disease conclusions can be pooled as one effect.
Cancer odds, T1D autoimmunity timing, metabolic metabolites, Himalayan lifestyle geography, and mouse OMV colitis are different outcomes in different designs. Pooling them as a single microbiome effect erases those limits.
Related
Claim ledger
What the evidence shows
Drawn from 15 studies in this library. Mix labels say which citation roles are present; they are not a strength score. Supports means evidence for a finding; Challenges means evidence against a stated position; Qualifies marks scope.
Genetic instruments for gut taxa link a small set of microbiome–cancer associations under Mendelian randomisation.
Using Mendelian randomisation on gut-taxa genetic instruments, eleven stringent microbiome-to-cancer associations were reported (for example higher breast-cancer odds with Actinobacteria/Bifidobacteriales signals; Tyzzerella3 linked with higher lung adenocarcinoma but lower colorectal cancer risk).
- B. dorei rises before T1D autoimmunity— different disease and design — observational T1D timing, not cancer MR
- Gut microbes, blood metabolites, and metabolic traits— different outcome — metabolic traits/metabolites, not cancer
Study Role Design N Population Outcome Mendelian randomisation associates specific gut taxa with several cancers Supports Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses) B. dorei rises before T1D autoimmunity Qualifiesdifferent disease and design — observational T1D timing, not cancer MR 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 Gut microbes, blood metabolites, and metabolic traits Qualifiesdifferent outcome — metabolic traits/metabolites, not cancer Cross-sectionalMETSIM subset microbiome–metabolite–metabolic trait association analysis N=531 · 531 middle-aged Finnish men from the METSIM cohort (parent cohort 10,197) Middle-aged Finnish men in METSIM Associations of gut microbiota with plasma metabolites and metabolic traits Before type 1 diabetes autoimmunity, Bacteroides dorei can dominate children’s gut communities.
In children who later developed type 1 diabetes autoimmunity, Bacteroides dorei became dominant in the gut community before persistent autoantibodies (mean diagnosis age about 16.8 months).
- Mendelian randomisation associates specific gut taxa with several cancers— different disease and method — cancer MR in adults, not pediatric T1D timing
- Gut microbes, blood metabolites, and metabolic traits— different population — middle-aged metabolic traits, not autoimmunity onset
Study Role Design N Population Outcome 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 Mendelian randomisation associates specific gut taxa with several cancers Qualifiesdifferent disease and method — cancer MR in adults, not pediatric T1D timing Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses) Gut microbes, blood metabolites, and metabolic traits Qualifiesdifferent population — middle-aged metabolic traits, not autoimmunity onset Cross-sectionalMETSIM subset microbiome–metabolite–metabolic trait association analysis N=531 · 531 middle-aged Finnish men from the METSIM cohort (parent cohort 10,197) Middle-aged Finnish men in METSIM Associations of gut microbiota with plasma metabolites and metabolic traits In middle-aged men, gut composition associates with metabolites and metabolic-syndrome traits.
In middle-aged men, gut microbiota composition was associated with plasma metabolites and metabolic-syndrome–related traits, placing the microbiome alongside genetics and lifestyle as a correlated host factor.
- Mendelian randomisation associates specific gut taxa with several cancers— different outcome — cancer odds via MR, not metabolic metabolites
- Selective bacterial BSH shifts host metabolism— mechanistic enzyme study — not a population metabolic association
Study Role Design N Population Outcome Gut microbes, blood metabolites, and metabolic traits Supports Cross-sectionalMETSIM subset microbiome–metabolite–metabolic trait association analysis N=531 · 531 middle-aged Finnish men from the METSIM cohort (parent cohort 10,197) Middle-aged Finnish men in METSIM Associations of gut microbiota with plasma metabolites and metabolic traits Mendelian randomisation associates specific gut taxa with several cancers Qualifiesdifferent outcome — cancer odds via MR, not metabolic metabolites Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses) Selective bacterial BSH shifts host metabolism Qualifiesmechanistic enzyme study — not a population metabolic association Animal / in-vitroBacteroidetes BSH selectivity screen plus B. thetaiotaomicron BT2086 deletion in gnotobiotic mice ~20 Bacteroidetes strains screened; monocolonization used 12 mice per Bt WT/KO group (8 GF controls) — no single primary N Human-gut Bacteroidetes strains and germ-free C57BL/6 mice Host metabolic phenotypes after selective bile-salt hydrolase loss Bacterial bile-salt enzymes can change host bile acids and weight-related phenotypes in models.
Selective bacterial bile-salt hydrolases can change host bile-acid chemistry and weight-related phenotypes in colonization models—evidence that specific microbial enzymes, not only community membership lists, can matter for host physiology.
- How much do gut bacterial gene contents differ?— different question — strain gene-content variation across people
- Antibiotics hurt bee gut and survival— different host — honey bees, not mammalian colonization models
Study Role Design N Population Outcome Selective bacterial BSH shifts host metabolism Supports Animal / in-vitroBacteroidetes BSH selectivity screen plus B. thetaiotaomicron BT2086 deletion in gnotobiotic mice ~20 Bacteroidetes strains screened; monocolonization used 12 mice per Bt WT/KO group (8 GF controls) — no single primary N Human-gut Bacteroidetes strains and germ-free C57BL/6 mice Host metabolic phenotypes after selective bile-salt hydrolase loss How much do gut bacterial gene contents differ? Qualifiesdifferent question — strain gene-content variation across people 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 Antibiotics hurt bee gut and survival Qualifiesdifferent host — honey bees, not mammalian colonization models Animal / in-vitroWorker honeybees fed tetracycline; 16S community and survival vs germ-free controls Cup cages of 30 bees × 15 replicates per condition; 16S profiles n≈14–15 per arm — no single primary N Worker honeybees with conventional or germ-free guts Core gut microbiota disruption and mortality after tetracycline Strain-level gene content varies a lot within the same gut species — species labels are coarse.
Even within the same named gut species, strain-level gene content varies substantially among healthy people—so species labels alone understate functional diversity.
- Is the gut microbiome the same along the intestine?— different scale — mucosal vs luminal niches along the gut
- Human body microbiome biogeography— different scale — body-site biogeography, not within-species gene content
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 Is the gut microbiome the same along the intestine? Qualifiesdifferent scale — mucosal vs luminal niches along the gut Animal / in-vitro16S profiling of paired luminal and mucosa-associated communities plus bile-acid and SCFA measurements N=24 · 144 luminal and mucosal samples from 24 pigs Healthy swine intestinal tract sections Mucosal and luminal microbiomes differ by gut section in diversity, co-occurrence, and bile-acid/SCFA environments Human body microbiome biogeography Qualifiesdifferent scale — body-site biogeography, not within-species gene content 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 Himalayan gut communities tracked farming history, with foragers most distinct from Americans and recent farmers in between.
In 54 Himalayan people plus US comparators, stool 16S composition tracked agricultural dependence: Chepang foragers retained forager-associated taxa, Tharu farmers (centuries of agriculture) looked more American-like, and Raute/Raji (farming 30–40 years) sat in between—showing community divergence can appear within a generation of lifestyle change.
- Mendelian randomisation associates specific gut taxa with several cancers— cancer Mendelian randomisation in GWAS ancestries ≠ Himalayan lifestyle-gradient 16S geography
- B. dorei rises before T1D autoimmunity— pediatric T1D timing of B. dorei dominance ≠ adult foraging-to-farming community shift
Study Role Design N Population Outcome Does leaving foraging change the gut microbiome? Supports Cross-sectionalStool 16S comparison of four Himalayan groups along a foraging-to-farming gradient plus 10 European-descent Americans N=54 · 54 Himalayan participants (Chepang 14, Raji 9, Raute 11, Tharu 20); plus 10 Americans (N=64 in some genus analyses) Rural Himalayan Chepang, Raji, Raute, and Tharu communities compared with Americans of European descent Gut bacterial community composition and how agricultural dependence tracks divergence from a foraging microbiome Mendelian randomisation associates specific gut taxa with several cancers Qualifiescancer Mendelian randomisation in GWAS ancestries ≠ Himalayan lifestyle-gradient 16S geography Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses) B. dorei rises before T1D autoimmunity Qualifiespediatric T1D timing of B. dorei dominance ≠ adult foraging-to-farming community shift 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 Non-viable probiotic vesicles can still change experimental colitis in mice.
Oral outer-membrane vesicles from E. coli Nissle 1917 reduced DSS colitis severity in mice (less weight loss/DAI and oedema, better histology), cut IL-1β/TNF-α/IL-17 mRNA, raised IL-10, and restored TFF-3, without reversing ZO-1 downregulation.
- Does leaving foraging change the gut microbiome?— mouse DSS pretreatment with OMVs ≠ human Himalayan lifestyle microbiome geography
- Mendelian randomisation associates specific gut taxa with several cancers— experimental colitis cytokines ≠ human cancer odds from taxa instruments
Study Role Design N Population Outcome Can E. coli Nissle vesicles calm experimental colitis? Supports Animal / in-vitroDSS colitis in male C57BL/6J mice; 10-day EcN OMV pretreatment, 5-day 3% DSS, then recovery; oral OMVs 5 μg/day N=24 · non-colitic n=6, colitic n=9, OMV-treated n=9 (male C57BL/6J, 7–9 weeks) Male C57BL/6J mice with DSS-induced colitis DAI/weight, colon oedema, histology, cytokines, and barrier marker TFF-3 Does leaving foraging change the gut microbiome? Qualifiesmouse DSS pretreatment with OMVs ≠ human Himalayan lifestyle microbiome geography Cross-sectionalStool 16S comparison of four Himalayan groups along a foraging-to-farming gradient plus 10 European-descent Americans N=54 · 54 Himalayan participants (Chepang 14, Raji 9, Raute 11, Tharu 20); plus 10 Americans (N=64 in some genus analyses) Rural Himalayan Chepang, Raji, Raute, and Tharu communities compared with Americans of European descent Gut bacterial community composition and how agricultural dependence tracks divergence from a foraging microbiome Mendelian randomisation associates specific gut taxa with several cancers Qualifiesexperimental colitis cytokines ≠ human cancer odds from taxa instruments Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses) CRC tumor mucosa had lower 16S diversity, more Fusobacteria, and less Proteobacteria than healthy gut.
In 31 CRC patients vs 30 healthy volunteers, Shannon diversity was 3.43 vs 4.01 (P<0.001). Tumor mucosa had more Firmicutes (63.46% vs 43.36%) and Fusobacteria (10.58% vs 0.03%) and less Proteobacteria (10.66% vs 60.35%). Tumors vs adjacent mucosa: Lactococcus 50.85% vs 25.35%.
- Mendelian randomisation associates specific gut taxa with several cancers— 16S case-control of tumor mucosa ≠ GWAS-instrument MR of taxa vs cancer odds
Study Role Design N Population Outcome Is the colon lining’s microbiome different in cancer? Supports Case-control16S V3 pyrosequencing of CRC mucosa vs adjacent tissue vs healthy colonoscopy controls, including proximal vs distal tumors N=61 · 31 CRC patients (31 tumor, 20 adjacent; 15 proximal, 16 distal) plus 30 healthy volunteers (15 proximal, 15 distal mucosa); 81 sequenced samples Recently diagnosed CRC patients and healthy colonoscopy volunteers (Shanghai Sixth People's Hospital) Mucosa-associated 16S community structure, diversity, and taxon abundances Mendelian randomisation associates specific gut taxa with several cancers Qualifies16S case-control of tumor mucosa ≠ GWAS-instrument MR of taxa vs cancer odds Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses) Two-sample MR links specific gut taxa to postpartum hemorrhage and other pregnancy outcomes.
MR of 18,340 MiBioGen participants versus FinnGen pregnancy GWAS found taxon-specific IVW signals—e.g. Deltaproteobacteria raising PPH risk (OR 1.26) and Butyricimonas lowering it (OR 0.79); Holdemanella OR 0.41 for abruptio placentae. All F statistics were >10.
- Mendelian randomisation associates specific gut taxa with several cancers— pregnancy-outcome MR ≠ cancer-odds MR; both depend on instrument assumptions
- Is the colon lining’s microbiome different in cancer?— CRC mucosa 16S case-control ≠ genetic instruments for pregnancy GWAS
Study Role Design N Population Outcome Can gut microbes cause pregnancy complications? Supports Mendelian randomisationTwo-sample MR of MiBioGen gut-microbiome GWAS vs FinnGen R5 adverse pregnancy outcomes, IVW primary with MR-Egger/MR-PRESSO N=18340 · MiBioGen exposure GWAS: 18,340 participants, 24 cohorts, 196 taxa after dropping unnamed taxa; FinnGen R5 outcomes include PPH 3,670/98,626 and GDM 5,687/117,892 Host-genetics–microbiome GWAS (MiBioGen) and Finnish pregnancy-complication GWAS (FinnGen R5) IVW odds ratios for postpartum hemorrhage, abruptio placentae, spontaneous abortion, PROM, GDM, preeclampsia/eclampsia, and preterm birth Mendelian randomisation associates specific gut taxa with several cancers Qualifiespregnancy-outcome MR ≠ cancer-odds MR; both depend on instrument assumptions Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses) Is the colon lining’s microbiome different in cancer? QualifiesCRC mucosa 16S case-control ≠ genetic instruments for pregnancy GWAS Case-control16S V3 pyrosequencing of CRC mucosa vs adjacent tissue vs healthy colonoscopy controls, including proximal vs distal tumors N=61 · 31 CRC patients (31 tumor, 20 adjacent; 15 proximal, 16 distal) plus 30 healthy volunteers (15 proximal, 15 distal mucosa); 81 sequenced samples Recently diagnosed CRC patients and healthy colonoscopy volunteers (Shanghai Sixth People's Hospital) Mucosa-associated 16S community structure, diversity, and taxon abundances
Debates
Tensions and limits
Some items are genuine disagreements on the same question. Others mark different assays, populations, or outcomes.
Cancer MR, pediatric T1D timing, and adult metabolic-trait associations all involve gut microbiota, but they answer different disease and design questions. Treating them as one interchangeable “gut microbiome disease effect” collapses distinct estimands.
Cancer MR, pediatric T1D timing, and adult metabolic-trait associations all involve gut microbiota, but they answer different disease and design questions. Treating them as one interchangeable “gut microbiome disease effect” collapses distinct estimands.
- Mendelian randomisation associates specific gut taxa with several cancers
- B. dorei rises before T1D autoimmunity
- Gut microbes, blood metabolites, and metabolic traits
Study Role Design N Population Outcome Mendelian randomisation associates specific gut taxa with several cancers Supports Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses) 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 Gut microbes, blood metabolites, and metabolic traits Supports Cross-sectionalMETSIM subset microbiome–metabolite–metabolic trait association analysis N=531 · 531 middle-aged Finnish men from the METSIM cohort (parent cohort 10,197) Middle-aged Finnish men in METSIM Associations of gut microbiota with plasma metabolites and metabolic traits
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
Human observational and MR studies do not automatically generalise to bee antibiotic/pesticide disruptions or to enzyme-focused colonization experiments. Animal and insect systems test different host–microbe questions; they limit how far a human association travels rather than falsifying it.
Human observational and MR studies do not automatically generalise to bee antibiotic/pesticide disruptions or to enzyme-focused colonization experiments. Animal and insect systems test different host–microbe questions; they limit how far a human association travels rather than falsifying it.
- Antibiotics hurt bee gut and survival
- Pesticides reshape honey bee gut microbes
- Selective bacterial BSH shifts host metabolism
- Mendelian randomisation associates specific gut taxa with several cancers
Study Role Design N Population Outcome Antibiotics hurt bee gut and survival Supports Animal / in-vitroWorker honeybees fed tetracycline; 16S community and survival vs germ-free controls Cup cages of 30 bees × 15 replicates per condition; 16S profiles n≈14–15 per arm — no single primary N Worker honeybees with conventional or germ-free guts Core gut microbiota disruption and mortality after tetracycline Pesticides reshape honey bee gut microbes Supports Animal / in-vitroIn-hive pesticide exposures across sites; 16S and ITS pyrosequencing of foragers and brood N=24 · 24 sequenced samples (4 treatments × 3 sites × 2 bee types), each pooled from five bees Honey bees (Apis mellifera) exposed to in-hive pesticides Gut bacterial and fungal community composition under pesticide treatments Selective bacterial BSH shifts host metabolism Supports Animal / in-vitroBacteroidetes BSH selectivity screen plus B. thetaiotaomicron BT2086 deletion in gnotobiotic mice ~20 Bacteroidetes strains screened; monocolonization used 12 mice per Bt WT/KO group (8 GF controls) — no single primary N Human-gut Bacteroidetes strains and germ-free C57BL/6 mice Host metabolic phenotypes after selective bile-salt hydrolase loss Mendelian randomisation associates specific gut taxa with several cancers Supports Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses)
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
A Himalayan lifestyle-gradient community shift, mouse OMV rescue of DSS colitis, CRC mucosa 16S, and pregnancy-outcome MR all involve gut microbes, but they do not test the same host, exposure, or outcome as cancer MR or pediatric T1D timing studies.
A Himalayan lifestyle-gradient community shift, mouse OMV rescue of DSS colitis, CRC mucosa 16S, and pregnancy-outcome MR all involve gut microbes, but they do not test the same host, exposure, or outcome as cancer MR or pediatric T1D timing studies.
- Does leaving foraging change the gut microbiome?
- Can E. coli Nissle vesicles calm experimental colitis?
- Mendelian randomisation associates specific gut taxa with several cancers
- Is the colon lining’s microbiome different in cancer?
- Can gut microbes cause pregnancy complications?
Study Role Design N Population Outcome Does leaving foraging change the gut microbiome? Supports Cross-sectionalStool 16S comparison of four Himalayan groups along a foraging-to-farming gradient plus 10 European-descent Americans N=54 · 54 Himalayan participants (Chepang 14, Raji 9, Raute 11, Tharu 20); plus 10 Americans (N=64 in some genus analyses) Rural Himalayan Chepang, Raji, Raute, and Tharu communities compared with Americans of European descent Gut bacterial community composition and how agricultural dependence tracks divergence from a foraging microbiome Can E. coli Nissle vesicles calm experimental colitis? Supports Animal / in-vitroDSS colitis in male C57BL/6J mice; 10-day EcN OMV pretreatment, 5-day 3% DSS, then recovery; oral OMVs 5 μg/day N=24 · non-colitic n=6, colitic n=9, OMV-treated n=9 (male C57BL/6J, 7–9 weeks) Male C57BL/6J mice with DSS-induced colitis DAI/weight, colon oedema, histology, cytokines, and barrier marker TFF-3 Mendelian randomisation associates specific gut taxa with several cancers Supports Mendelian randomisationTwo-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets 211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia) Causal microbiome–cancer associations (stringent and sensitivity analyses) Is the colon lining’s microbiome different in cancer? Supports Case-control16S V3 pyrosequencing of CRC mucosa vs adjacent tissue vs healthy colonoscopy controls, including proximal vs distal tumors N=61 · 31 CRC patients (31 tumor, 20 adjacent; 15 proximal, 16 distal) plus 30 healthy volunteers (15 proximal, 15 distal mucosa); 81 sequenced samples Recently diagnosed CRC patients and healthy colonoscopy volunteers (Shanghai Sixth People's Hospital) Mucosa-associated 16S community structure, diversity, and taxon abundances Can gut microbes cause pregnancy complications? Supports Mendelian randomisationTwo-sample MR of MiBioGen gut-microbiome GWAS vs FinnGen R5 adverse pregnancy outcomes, IVW primary with MR-Egger/MR-PRESSO N=18340 · MiBioGen exposure GWAS: 18,340 participants, 24 cohorts, 196 taxa after dropping unnamed taxa; FinnGen R5 outcomes include PPH 3,670/98,626 and GDM 5,687/117,892 Host-genetics–microbiome GWAS (MiBioGen) and Finnish pregnancy-complication GWAS (FinnGen R5) IVW odds ratios for postpartum hemorrhage, abruptio placentae, spontaneous abortion, PROM, GDM, preeclampsia/eclampsia, and preterm birth
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
Timeline
How understanding moved
Study years are when the paper was published. Evidence edits are dated changes to this page's claims. Explanations are when PaperFren added a Discovery — not a claim that the science happened that day.
2026
Concept page published
Microbiome
Added a misconception
Mendelian randomisation proving a taxa–cancer association means the microbe’s mechanism in tumours is known.
Added a misconception
B. dorei dominance before autoantibodies proves that B. dorei causes type 1 diabetes.
Added a misconception
If two papers both study “the gut microbiome,” their disease conclusions can be pooled as one effect.
- Added a scope qualifier
Antibiotics hurt bee gut and survival
- Added a scope qualifier
How much do gut bacterial gene contents differ?
- Added a scope qualifier
B. dorei rises before T1D autoimmunity
- Added a scope qualifier
Gut microbes, blood metabolites, and metabolic traits
- Added a scope qualifier
Mendelian randomisation associates specific gut taxa with several cancers
- Added a scope qualifier
Gut microbes, blood metabolites, and metabolic traits
- Added a scope qualifier
Selective bacterial BSH shifts host metabolism
- Added a scope qualifier
Mendelian randomisation associates specific gut taxa with several cancers
- Added a scope qualifier
Human body microbiome biogeography
- Added a scope qualifier
Is the gut microbiome the same along the intestine?
- Added a supporting study
Mendelian randomisation associates specific gut taxa with several cancers
- Added a supporting study
B. dorei rises before T1D autoimmunity
- Added a supporting study
Gut microbes, blood metabolites, and metabolic traits
- Added a supporting study
How much do gut bacterial gene contents differ?
- Added a supporting study
B. dorei rises before T1D autoimmunity
- Added a supporting study
Mendelian randomisation associates specific gut taxa with several cancers
- Added a supporting study
Gut microbes, blood metabolites, and metabolic traits
- Added a supporting study
Selective bacterial BSH shifts host metabolism
- Added a supporting study
Antibiotics hurt bee gut and survival
- Added a supporting study
Mendelian randomisation associates specific gut taxa with several cancers
- Added a supporting study
Pesticides reshape honey bee gut microbes
Claim withdrawn
Among ~20 Bacteroidetes strains, some BSHs prefer steroid cores; BtΔ2086 loses deconjugation; colonization changes weight on high-fat diet vs WT.
Claim withdrawn
Cases show B. dorei dominance prior to persistent autoantibodies (mean diagnosis age 16.8 months).
Claim withdrawn
Strain-level gene content varies substantially among individuals for the same species in natural habitats.
Claim withdrawn
Four of eight core gut species dropped; elevated mortality in treated bees; germ-free bees did not show the same antibiotic mortality spike.
New claim
Using Mendelian randomisation on gut-taxa genetic instruments, eleven stringent microbiome-to-cancer associations were reported (for example higher breast-cancer odds with Actinobacteria/Bifidobacteriales signals; Tyzzerella3 linked with higher lung adenocarcinoma but lower colorectal cancer risk).
New claim
In children who later developed type 1 diabetes autoimmunity, Bacteroides dorei became dominant in the gut community before persistent autoantibodies (mean diagnosis age about 16.8 months).
New claim
In middle-aged men, gut microbiota composition was associated with plasma metabolites and metabolic-syndrome–related traits, placing the microbiome alongside genetics and lifestyle as a correlated host factor.
New claim
Even within the same named gut species, strain-level gene content varies substantially among healthy people—so species labels alone understate functional diversity.
New tension
Cancer MR, pediatric T1D timing, and adult metabolic-trait associations all involve gut microbiota, but they answer different disease and design questions. Treating them as one interchangeable “gut microbiome disease effect” collapses distinct estimands.
New tension
Human observational and MR studies do not automatically generalise to bee antibiotic/pesticide disruptions or to enzyme-focused colonization experiments. Animal and insect systems test different host–microbe questions; they limit how far a human association travels rather than falsifying it.
Removed a misconception
Microbiome findings from one model organism always transfer to humans.
- Removed a supporting study
B. dorei rises before T1D autoimmunity
- Removed a supporting study
How much do gut bacterial gene contents differ?
Tension withdrawn
Systems and scales differ across microbiome studies (species, tissues, methods), so mechanisms should not be over-generalised.
- Added a scope qualifier
B. dorei rises before T1D autoimmunity
- Added a scope qualifier
Mendelian randomisation associates specific gut taxa with several cancers
- Added a scope qualifier
Does leaving foraging change the gut microbiome?
- Added a scope qualifier
Mendelian randomisation associates specific gut taxa with several cancers
- Added a supporting study
Does leaving foraging change the gut microbiome?
- Added a supporting study
Can E. coli Nissle vesicles calm experimental colitis?
- Added a supporting study
Does leaving foraging change the gut microbiome?
- Added a supporting study
Mendelian randomisation associates specific gut taxa with several cancers
- Added a supporting study
Can E. coli Nissle vesicles calm experimental colitis?
New claim
In 54 Himalayan people plus US comparators, stool 16S composition tracked agricultural dependence: Chepang foragers retained forager-associated taxa, Tharu farmers (centuries of agriculture) looked more American-like, and Raute/Raji (farming 30–40 years) sat in between—showing community divergence can appear within a generation of lifestyle change.
New claim
Oral outer-membrane vesicles from E. coli Nissle 1917 reduced DSS colitis severity in mice (less weight loss/DAI and oedema, better histology), cut IL-1β/TNF-α/IL-17 mRNA, raised IL-10, and restored TFF-3, without reversing ZO-1 downregulation.
New tension
A Himalayan lifestyle-gradient community shift and mouse OMV rescue of DSS colitis both involve gut microbes, but they do not test the same host, exposure, or outcome as cancer MR or pediatric T1D timing studies.
- Added a scope qualifier
Mendelian randomisation associates specific gut taxa with several cancers
- Added a scope qualifier
Mendelian randomisation associates specific gut taxa with several cancers
- Added a scope qualifier
Is the colon lining’s microbiome different in cancer?
- Added a supporting study
Is the colon lining’s microbiome different in cancer?
- Added a supporting study
Can gut microbes cause pregnancy complications?
- Added a supporting study
Can gut microbes cause pregnancy complications?
- Added a supporting study
Is the colon lining’s microbiome different in cancer?
New claim
In 31 CRC patients vs 30 healthy volunteers, Shannon diversity was 3.43 vs 4.01 (P<0.001). Tumor mucosa had more Firmicutes (63.46% vs 43.36%) and Fusobacteria (10.58% vs 0.03%) and less Proteobacteria (10.66% vs 60.35%). Tumors vs adjacent mucosa: Lactococcus 50.85% vs 25.35%.
New claim
MR of 18,340 MiBioGen participants versus FinnGen pregnancy GWAS found taxon-specific IVW signals—e.g. Deltaproteobacteria raising PPH risk (OR 1.26) and Butyricimonas lowering it (OR 0.79); Holdemanella OR 0.41 for abruptio placentae. All F statistics were >10.
Change log
What changed
Dated edits to this page's evidence: studies added or removed from a claim, claims added or withdrawn, and new explanations tagged here. Coverage edits are distinct from reviewed research changes and corrections. Dates record PaperFren updates; study publication dates appear in the timeline.
In 31 CRC patients vs 30 healthy volunteers, Shannon diversity was 3.43 vs 4.01 (P<0.001). Tumor mucosa had more Firmicutes (63.46% vs 43.36%) and Fusobacteria (10.58% vs 0.03%) and less Proteobacteria (10.66% vs 60.35%). Tumors vs adjacent mucosa: Lactococcus 50.85% vs 25.35%.
- Coverage updated · New claim
- Coverage updated · Added a supporting study: Is the colon lining’s microbiome different in cancer?
- Coverage updated · Added a scope qualifier: Mendelian randomisation associates specific gut taxa with several cancers
MR of 18,340 MiBioGen participants versus FinnGen pregnancy GWAS found taxon-specific IVW signals—e.g. Deltaproteobacteria raising PPH risk (OR 1.26) and Butyricimonas lowering it (OR 0.79); Holdemanella OR 0.41 for abruptio placentae. All F statistics were >10.
- Coverage updated · New claim
- Coverage updated · Added a supporting study: Can gut microbes cause pregnancy complications?
- Coverage updated · Added a scope qualifier: Mendelian randomisation associates specific gut taxa with several cancers
- Coverage updated · Added a scope qualifier: Is the colon lining’s microbiome different in cancer?
A Himalayan lifestyle-gradient community shift, mouse OMV rescue of DSS colitis, CRC mucosa 16S, and pregnancy-outcome MR all involve gut microbes, but they do not test the same host, exposure, or outcome as cancer MR or pediatric T1D timing studies.
- Coverage updated · Added a supporting study: Can gut microbes cause pregnancy complications?
- Coverage updated · Added a supporting study: Is the colon lining’s microbiome different in cancer?
In 54 Himalayan people plus US comparators, stool 16S composition tracked agricultural dependence: Chepang foragers retained forager-associated taxa, Tharu farmers (centuries of agriculture) looked more American-like, and Raute/Raji (farming 30–40 years) sat in between—showing community divergence can appear within a generation of lifestyle change.
- Coverage updated · New claim
- Coverage updated · Added a supporting study: Does leaving foraging change the gut microbiome?
- Coverage updated · Added a scope qualifier: B. dorei rises before T1D autoimmunity
- Coverage updated · Added a scope qualifier: Mendelian randomisation associates specific gut taxa with several cancers
Oral outer-membrane vesicles from E. coli Nissle 1917 reduced DSS colitis severity in mice (less weight loss/DAI and oedema, better histology), cut IL-1β/TNF-α/IL-17 mRNA, raised IL-10, and restored TFF-3, without reversing ZO-1 downregulation.
- Coverage updated · New claim
- Coverage updated · Added a supporting study: Can E. coli Nissle vesicles calm experimental colitis?
- Coverage updated · Added a scope qualifier: Does leaving foraging change the gut microbiome?
- Coverage updated · Added a scope qualifier: Mendelian randomisation associates specific gut taxa with several cancers
A Himalayan lifestyle-gradient community shift and mouse OMV rescue of DSS colitis both involve gut microbes, but they do not test the same host, exposure, or outcome as cancer MR or pediatric T1D timing studies.
- Coverage updated · New tension
- Coverage updated · Added a supporting study: Does leaving foraging change the gut microbiome?
- Coverage updated · Added a supporting study: Mendelian randomisation associates specific gut taxa with several cancers
- Coverage updated · Added a supporting study: Can E. coli Nissle vesicles calm experimental colitis?
Selective bacterial bile-salt hydrolases can change host bile-acid chemistry and weight-related phenotypes in colonization models—evidence that specific microbial enzymes, not only community membership lists, can matter for host physiology.
- Coverage updated · Removed a supporting study: B. dorei rises before T1D autoimmunity
- Coverage updated · Removed a supporting study: How much do gut bacterial gene contents differ?
- Coverage updated · Added a scope qualifier: Antibiotics hurt bee gut and survival
- Coverage updated · Added a scope qualifier: How much do gut bacterial gene contents differ?
Using Mendelian randomisation on gut-taxa genetic instruments, eleven stringent microbiome-to-cancer associations were reported (for example higher breast-cancer odds with Actinobacteria/Bifidobacteriales signals; Tyzzerella3 linked with higher lung adenocarcinoma but lower colorectal cancer risk).
- Coverage updated · New claim
- Coverage updated · Added a supporting study: Mendelian randomisation associates specific gut taxa with several cancers
- Coverage updated · Added a scope qualifier: B. dorei rises before T1D autoimmunity
- Coverage updated · Added a scope qualifier: Gut microbes, blood metabolites, and metabolic traits
In children who later developed type 1 diabetes autoimmunity, Bacteroides dorei became dominant in the gut community before persistent autoantibodies (mean diagnosis age about 16.8 months).
- Coverage updated · New claim
- Coverage updated · Added a supporting study: B. dorei rises before T1D autoimmunity
- Coverage updated · Added a scope qualifier: Mendelian randomisation associates specific gut taxa with several cancers
- Coverage updated · Added a scope qualifier: Gut microbes, blood metabolites, and metabolic traits
In middle-aged men, gut microbiota composition was associated with plasma metabolites and metabolic-syndrome–related traits, placing the microbiome alongside genetics and lifestyle as a correlated host factor.
- Coverage updated · New claim
- Coverage updated · Added a supporting study: Gut microbes, blood metabolites, and metabolic traits
- Coverage updated · Added a scope qualifier: Selective bacterial BSH shifts host metabolism
- Coverage updated · Added a scope qualifier: Mendelian randomisation associates specific gut taxa with several cancers
Even within the same named gut species, strain-level gene content varies substantially among healthy people—so species labels alone understate functional diversity.
- Coverage updated · New claim
- Coverage updated · Added a supporting study: How much do gut bacterial gene contents differ?
- Coverage updated · Added a scope qualifier: Human body microbiome biogeography
- Coverage updated · Added a scope qualifier: Is the gut microbiome the same along the intestine?
Among ~20 Bacteroidetes strains, some BSHs prefer steroid cores; BtΔ2086 loses deconjugation; colonization changes weight on high-fat diet vs WT.
- Coverage updated · Claim withdrawn
Show 11 earlierShow fewer
Cases show B. dorei dominance prior to persistent autoantibodies (mean diagnosis age 16.8 months).
- Coverage updated · Claim withdrawn
Strain-level gene content varies substantially among individuals for the same species in natural habitats.
- Coverage updated · Claim withdrawn
Four of eight core gut species dropped; elevated mortality in treated bees; germ-free bees did not show the same antibiotic mortality spike.
- Coverage updated · Claim withdrawn
Cancer MR, pediatric T1D timing, and adult metabolic-trait associations all involve gut microbiota, but they answer different disease and design questions. Treating them as one interchangeable “gut microbiome disease effect” collapses distinct estimands.
- Coverage updated · New tension
- Coverage updated · Added a supporting study: B. dorei rises before T1D autoimmunity
- Coverage updated · Added a supporting study: Mendelian randomisation associates specific gut taxa with several cancers
- Coverage updated · Added a supporting study: Gut microbes, blood metabolites, and metabolic traits
Human observational and MR studies do not automatically generalise to bee antibiotic/pesticide disruptions or to enzyme-focused colonization experiments. Animal and insect systems test different host–microbe questions; they limit how far a human association travels rather than falsifying it.
- Coverage updated · New tension
- Coverage updated · Added a supporting study: Selective bacterial BSH shifts host metabolism
- Coverage updated · Added a supporting study: Antibiotics hurt bee gut and survival
- Coverage updated · Added a supporting study: Mendelian randomisation associates specific gut taxa with several cancers
- Coverage updated · Added a supporting study: Pesticides reshape honey bee gut microbes
Systems and scales differ across microbiome studies (species, tissues, methods), so mechanisms should not be over-generalised.
- Coverage updated · Tension withdrawn
Mendelian randomisation proving a taxa–cancer association means the microbe’s mechanism in tumours is known.
- Coverage updated · Added a misconception
B. dorei dominance before autoantibodies proves that B. dorei causes type 1 diabetes.
- Coverage updated · Added a misconception
If two papers both study “the gut microbiome,” their disease conclusions can be pooled as one effect.
- Coverage updated · Added a misconception
Microbiome findings from one model organism always transfer to humans.
- Coverage updated · Removed a misconception
- Coverage updated · Concept page published
Papers
15 studies in this library bear on Microbiome, ordered by citations.
- Human body microbiome biogeography
Healthy human body sites host distinct bacterial communities across a massive 16S survey.
- Is the colon lining’s microbiome different in cancer?
In 31 CRC patients vs 30 healthy volunteers, tumor mucosa had lower 16S diversity, more Firmicutes and Fusobacteria, and less Proteobacteria; Lactococcus/Fusobacterium rose in tumors while Pseudomonas fell.
- Antibiotics hurt bee gut and survival
Tetracycline shrinks the honeybee gut microbiome and elevates mortality after hive return.
- Mendelian randomisation associates specific gut taxa with several cancers
Genetic instruments for gut microbiota taxa showed causal associations with breast, lung, colorectal, prostate, gastric, and head/neck cancers—sometimes in opposing directions for the same genus.
- Selective bacterial BSH shifts host metabolism
Bacteroides bile salt hydrolase BT2086 selectively deconjugates bile acids and alters host metabolism.
- Gut microbes, blood metabolites, and metabolic traits
In 531 METSIM men, gut microbiota associate with plasma metabolites and metabolic-syndrome-relevant traits.
- B. dorei rises before T1D autoimmunity
Bacteroides dorei dominates the gut microbiome before autoimmunity in high-risk Finnish children.
- Can E. coli Nissle vesicles calm experimental colitis?
Oral outer-membrane vesicles from probiotic E. coli Nissle 1917 reduced DSS colitis severity in mice and lowered inflammatory cytokines while restoring TFF-3.
Show 7 more studiesShow fewer studies
- How do gut-bacteria vesicles enter host cells?
B. thetaiotaomicron outer-membrane vesicles are taken up within minutes via dynamin-dependent endocytosis/macropinocytosis and traffic to Golgi/ER/lysosomes.
- Infant diet shapes gut–host immunity
Breast- vs formula-fed infants differ in gut microbiota and host immune gene expression.
- Pesticides reshape honey bee gut microbes
In-hive pesticide exposures change honey bee gut bacterial and fungal community composition.
- How much do gut bacterial gene contents differ?
Same gut bacterial species can differ widely in gene content across people, shaping functional individuality.
- Is the gut microbiome the same along the intestine?
Swine mucosal and luminal microbiomes differ by gut section, with distinct diversity patterns and bile-acid/SCFA environments.
- Does leaving foraging change the gut microbiome?
Among 54 Himalayan people, gut microbiomes tracked farming history: Chepang foragers looked most traditional, Tharu farmers closer to Americans, and recently transitioned Raute/Raji sat in between.
- Can gut microbes cause pregnancy complications?
Two-sample MR of 18,340 MiBioGen participants versus FinnGen pregnancy GWAS found taxon-specific IVW signals—e.g. Deltaproteobacteria raising PPH risk (OR 1.26) and Butyricimonas lowering it (OR 0.79).
Compare studies
Select 2–10 studies. Design and N are labels, not a ranking.
Nothing selected yet.
Questions
What is still open
Cancer MR, pediatric T1D timing, and adult metabolic-trait associations all involve gut microbiota, but they answer different disease and design questions. Treating them as one interchangeable “gut microbiome disease effect” collapses distinct estimands.
Human observational and MR studies do not automatically generalise to bee antibiotic/pesticide disruptions or to enzyme-focused colonization experiments. Animal and insect systems test different host–microbe questions; they limit how far a human association travels rather than falsifying it.
A Himalayan lifestyle-gradient community shift, mouse OMV rescue of DSS colitis, CRC mucosa 16S, and pregnancy-outcome MR all involve gut microbes, but they do not test the same host, exposure, or outcome as cancer MR or pediatric T1D timing studies.
Ask PaperFren about Microbiome
Study this conceptflashcards and short-answer questions
A classmate says the MR cancer paper and the B. dorei T1D paper disagree about whether the microbiome is harmful. How do you respond?
They are not answering the same question. One reports genetic-instrument associations with specific cancers; the other reports a temporal dominance pattern before T1D autoimmunity. Different diseases and designs are a scope limit, not a forced disagreement about a single “harm” effect.
What would most strengthen a causal position that a named gut taxon increases a cancer’s risk?
Convergent evidence beyond MR assumptions—such as strain-resolved exposures, experimental models that recover the pathway, and replication outside the original ancestry GWAS—while keeping the cancer endpoint specific.
Why does strain-level gene-content variation qualify species-level microbiome claims?
People can share a species name while differing in gene content, so functions (for example enzyme capacity) may not travel with the species label alone.
A classmate pools the Himalaya foraging–farming stool study with Nissle OMV colitis mice as proof that “traditional microbiomes prevent IBD.” What is wrong?
The human paper is a cross-sectional 16S geography of lifestyle, not IBD incidence. The mouse paper is experimental DSS colitis with pretreatment vesicles. Neither estimates human IBD risk from a foraging microbiome.
Flashcards
0 of 8 answers reviewed