Bacterial strains in the human gut associate with host physiology

Saar Shoer, Anastasia Godneva, Adina Weinberger, Katherine S. Pollard, Yitzhak Pilpel, Eran SegalbioRxiv 2025 · DOI

Abstract

The human microbiota influences host physiology, yet much of its complexity lies beneath the species level. Here, we analyzed the intra-species genetic diversity of 936 gut bacteria across 24,997 individuals from three countries. Our findings show that highly abundant species exhibit greater strain stability, whereas low-abundance species display increased in-clonal mutations. Clonal strains are often mutually exclusive, while genetically variable strains tend to coexist. Strain turnover is associated with the presence of annotated chemotaxis and sporulation genes in reference genomes, whereas strain coexistence is associated with quorum sensing and secretion systems. Leveraging deep phenotypic data, we constructed an atlas detailing strain-level associations with diverse host physiological domains. For example, Phocaeicola vulgatus sub-types relate to host obesity, Lachnospira eligens to sleep, and Parabacteroides distasonis to iron hemostasis. This resource may guide personalized microbiome-based interventions to improve human health.

Introduction

Most microbiome research treats bacterial species as monolithic entities. However, just as two humans can have vastly different traits despite belonging to the same species, two strains of the same bacterium can behave differently in the gut. The traditional "species-level" view is often too coarse to explain the nuanced relationships between the microbiome and host health. Until now, we lacked a large-scale understanding of how intra-species genetic diversity—the subtle genomic variations within a single species—relates to human physiology. This study addresses that gap by moving from a "who is there" (species) approach to a "what version of them is there" (strain) approach across a massive, multi-country population.

Methods and Datasets

The researchers utilized metagenomic sequencing to analyze Single Nucleotide Polymorphisms (SNPs) across 29,586 fecal samples from 24,997 individuals. The study integrated data from four major sources:

  1. The Israeli Human Phenotype Project: The primary discovery cohort with deep longitudinal data.
  2. Lifelines (Netherlands): A validation cohort.
  3. DayTwo (United States): A second validation cohort.
  4. The "Ullman" Cohort: High-resolution longitudinal data from a single individual (86 samples over two years) to track daily strain dynamics.

Key metrics included SNP dissimilarity and Linkage Disequilibrium (LD) to measure clonality. Machine learning models (XGBoost) were trained to predict bacterial sub-types based on thousands of clinical and behavioral phenotypes.

Key Insights into Strain Dynamics

Stability and Mutation

The study found that highly abundant species tend to have very stable strains. In contrast, low-abundance species show more "in-clonal" mutations, likely due to higher selective pressures or smaller population sizes. Furthermore, clonal strains (those with low recombination) are usually mutually exclusive within a host, whereas genetically variable strains are more likely to coexist with other versions of themselves.

The Mechanics of Turnover

Strain turnover—the replacement of one strain by another—was linked to genes for chemotaxis and flagella. This suggests that motile bacteria are better equipped to colonize or leave the host environment. Conversely, strain coexistence was associated with quorum sensing and secretion systems, indicating a more "social" or cooperative survival strategy.

Microbial Fingerprinting

Bacterial strains serve as a more reliable "fingerprint" than species composition. Strain typing could identify an individual after four years with over 94% accuracy, whereas species abundance profiles were successful in less than 30% of cases.

Results: An Atlas of Host-Microbe Associations

The researchers constructed an atlas of 88 significant associations between bacterial sub-types and host physiology. These associations often revealed patterns that were invisible at the species level:

  • Obesity: Specific sub-types of Phocaeicola vulgatus (formerly Bacteroides vulgatus) were strongly linked to Body Mass Index (BMI). Different strains within this species sometimes showed opposite effects on host weight.
  • Sleep: Lachnospira eligens sub-types were associated with sleep apnea indicators and overall sleep quality.
  • Iron Homeostasis: Parabacteroides distasonis strains were tied to markers of iron levels, such as ferritin and hemoglobin.
  • Environmental Dominance: Cohabiting partners shared significantly more strains (25%) than siblings who lived apart (6.45%), proving that shared environment shapes the strain profile more than host genetics.

Limitations and Future Directions

While this study provides a high-resolution map of the gut microbiome, it is primarily observational and focused on Western populations (Israel, Netherlands, and the US). Future work must establish causality and explore these dynamics in non-Western cohorts with different diets and lifestyles. The authors suggest that these findings pave the way for personalized microbiome-based interventions, where specific disease-linked strains could be targeted or replaced with health-promoting alternatives.

Summary updated

Abstract

The human microbiota influences host physiology, yet much of its complexity lies beneath the species level. Here, we analyzed the intra-species genetic diversity of 936 gut bacteria across 24,997 individuals from three countries. Our findings show that highly abundant species exhibit greater strain stability, whereas low-abundance species display increased in-clonal mutations. Clonal strains are often mutually exclusive, while genetically variable strains tend to coexist. Strain turnover is associated with the presence of annotated chemotaxis and sporulation genes in reference genomes, whereas strain coexistence is associated with quorum sensing and secretion systems. Leveraging deep phenotypic data, we constructed an atlas detailing strain-level associations with diverse host physiological domains. For example, <i>Phocaeicola vulgatus</i> sub-types relate to host obesity, <i>Lachnospira eligens</i> to sleep, and <i>Parabacteroides distasonis</i> to iron hemostasis. This resource may guide personalized microbiome-based interventions to improve human health.
BibTeX
@misc{shoer2025bacterial,
  author = {Saar Shoer and Anastasia Godneva and Adina Weinberger and Katherine S. Pollard and Yitzhak Pilpel and Eran Segal},
  title = {{Bacterial strains in the human gut associate with host physiology}},
  year = {2025},
  doi = {10.1101/2025.11.30.691432},
  howpublished = {bioRxiv},
}
Figure