Five microbiome studies that shaped the past year

Five microbiome studies that shaped the past year

By Alena Pribyl, Lead Scientist at Microba

The microbiome field continues to mature from descriptive cataloguing toward quantitative, mechanistic, and clinically actionable science. The five studies I’ve picked below stood out to me over the past year because each one tackles a question that has been hanging over the field for years — what does diet actually do to the microbiome at species resolution, how long do antibiotics really linger, what are we missing by only studying culturable bugs, how much energy does the microbiome really give us, and how predictable are community responses to drug perturbations? Here’s why I think each one matters.

1. Segev et al., 2026¹ — Nature Medicine

Diet–microbiome associations in 10,068 individuals from the Human Phenotype Project to guide personalized nutrition

Using app-based food logs and shotgun metagenomics from over 10,000 deeply phenotyped individuals, the authors built predictive models linking diet to species-level microbiome composition, diversity, and microbial pathways, and showed these associations persist over four years.

Why this matters: This landmark nutrition-microbiome study shows that food choices are strongly linked with the gut microbiome. Diets higher in minimally processed, nutrient-rich foods were linked with greater microbial diversity, while diets higher in ultra-processed foods were linked with lower diversity. Because the study included more than 10,000 people, used species-level microbiome testing, and followed some participants over time, it provides some of the strongest evidence to date that practical, food-based guidance is relevant to improving gut microbiome health.

2. Baldanzi et al., 2026² — Nature Medicine

Antibiotic use and gut microbiome composition links from individual-level prescription data of 14,979 individuals

By linking Sweden’s national prescription register to faecal metagenomes from nearly 15,000 adults, the authors quantified how oral antibiotic use over the preceding eight years tracks with gut microbiome composition at species level.

Why this matters: This work demonstrates that antibiotic use from up to eight years ago can still be associated with measurable changes in the microbiome, and identifies which antibiotic classes (clindamycin, fluoroquinolones, and flucloxacillin) carry the heaviest microbiome cost — adding a new dimension to antibiotic stewardship arguments beyond resistance alone.

3. da Silva et al., 2026³ — Cell Host & Microbe

Meta-analysis of the uncultured gut microbiome across 11,115 global metagenomes reveals a candidate signature of health

Across 11,115 metagenomes spanning 39 countries and 13 noncommunicable diseases, the authors found that uncultured bacteria are overrepresented among health-associated taxa, with the largely uncultured genus CAG-170 emerging as the strongest health-associated lineage globally.

Why this matters: This study highlights the importance of assessing both cultured and uncultured bacteria in the gut microbiome, and provides striking evidence that some of the strongest signatures of health come from understudied species that have largely escaped attention because they can’t (yet) be grown in a Petri dish.

4. Arnoldini et al., 2025⁴ — Cell

Quantifying the varying harvest of fermentation products from the human gut microbiota

Integrating bacterial metabolism, digestive physiology, and metagenomics, the authors put quantitative numbers on how much energy the gut microbiota delivers to its host across diets and species — 1.8–12.1% in humans (2–5% on Western diets, >10% on non-Western diets) versus >21% in laboratory mice.

Why this matters: This study provides a rigorous quantitative framework for estimating how much energy people get from gut bacteria breaking down food, especially fibre and other complex carbohydrates. It also reminds us to be careful when applying mouse microbiome findings to people, because mice appear to get a much larger share of their energy from gut bacterial fermentation than humans do. This means microbiome changes may have stronger metabolic effects in mice than they would in patients.

5. Shi et al., 2025⁵ — Cell

Nutrient competition predicts gut microbiome restructuring under drug perturbations

Screening stool-derived in vitro communities against 707 clinically relevant drugs across ~5,000 community–drug conditions, the authors showed that a species’ response in a community can largely be predicted from its monoculture drug susceptibility rescaled by the susceptibility of its competitors.

Why this matters: This study helps explain why medicines can cause unexpected shifts in the gut microbiome. When a drug suppresses some bacteria, other bacteria may grow because they face less competition for nutrients. This gives researchers a practical framework for predicting which microbiome changes may occur during treatment, and how they might eventually be reduced.

Together, these five studies move us closer to a microbiome science that is quantitative, predictive, and clinically translatable — whether the goal is personalising diet, stewarding antibiotics, redefining what “healthy” looks like at the species level, recalibrating how we interpret mouse data, or anticipating the collateral effects of the drugs we prescribe.