sciencebriefs
9:26in productionCh. 1 · A reference project, not a clinical one/ 9:26 · ceiling 15 min
Medicine · Genetics

Human Microbiome Project

It gave us the first map of where microbes live in healthy humans — not why they matter, not how to fix them, just where they are.

The Human Microbiome Project built the first large-scale reference database of human-associated microbes using culture-independent sequencing. It mapped microbial presence across body sites in healthy volunteers. It delivered essential infrastructure — not explanations, not cures, not definitions of health. Its value is in what it enables downstream, not what it concluded.

MedicineGeneticsDiseaseGeneticsMedicineDisease across the network →
Chapters & takeaways4
  1. 1:04
    A reference project, not a clinical one

    The HMP was a U.S. NIH initiative to map human microbiota — not treat disease, not test therapies, just build infrastructure.

  2. 2:30
    How the map was made

    It sequenced 16S rRNA genes from >5,000 samples across 15–18 body sites in 242 healthy people — all DNA run on sequencing machines.

  3. 3:56
    3000 isolates, not 3000 species

    It planned 3000 reference genomes from isolated bacteria — a benchmark for metagenomic analysis, not a complete species list.

  4. 5:40
    Two phases, two goals

    iHMP (2014–2016) shifted focus to health and disease states — but the original HMP1 (2007–2014) did not include disease cohorts.

Worth your time?

Yes. Study the whole thing.

4/ 5
What works
  • reference genome generation
  • multi-site microbial profiling
  • standardised 16S sequencing protocol
What does not
  • establish causality
  • define healthy microbiome
  • include disease cohorts
  • validate therapeutic interventions
Study it if
  • microbiome researchers
  • clinical trial designers
  • bioinformaticians
Skip it if
  • patients seeking microbiome-based treatments
  • clinicians without bioinformatics support
  • policy makers expecting immediate health applications
The written brief1 min read

What the work claims

The project claimed to improve understanding of human microbiota in health and disease by generating foundational resources: reference genomes, taxonomic profiles, and metagenomic benchmarks.

How it was done

The Human Microbiome Project sequenced bacterial 16S rRNA genes from over 5,000 samples across 15–18 body sites in 242 healthy U.S. volunteers. It generated 3000 reference genomic sequences from isolated bacterial strains. All human and microbial DNA was analysed using DNA sequencing machines.

What holds up

The reference database of 3000 bacterial isolate genomes holds up. The 16S rRNA sequence catalogue from 242 healthy individuals across multiple body sites holds up. The demonstration of functional redundancy and dynamic composition across body sites holds up.

What does not

It did not characterise microbiomes in disease. It did not establish causal mechanisms. It did not define ‘healthy’ microbiome composition. It did not assess longitudinal stability beyond snapshots.

Why it matters beyond the lab

It matters because every subsequent microbiome study in clinical trials, diagnostics, or therapeutics relies on its reference data to distinguish signal from noise — but only where that data applies.

Is it worth your time

Yes — it built the first large-scale, culture-independent reference framework for human-associated microbes. You need this baseline to interpret any microbiome study in health or disease.

Same field · Medicine4 of 84
13:00
COVID-19 pandemicSARS-CoV-2 was identified in Wuhan, China in late 2019, and the World Health Organization declared a pandemic on 11 March 2020, after which vaccines developed by groups including Pfizer-BioNTech, Moderna and Oxford-AstraZeneca reached the public from December 2020 onward at unusually high speed. Confirmed global deaths attributed to COVID-19 total in the low millions, but broader estimates using excess mortality, comparing how many more people died during the pandemic than would ordinarily be expected, run several times higher, reflecting substantial undercounting, particularly in places with limited testing and reporting capacity. The virus's genetic makeup points strongly to a natural, zoonotic origin involving bats, most likely through an intermediate animal host, and an early candidate explanation involving pangolins has not held up under closer genetic comparison, but the World Health Organization's own 2021 review found the precise pathway from animal reservoir to the first human infections remains unclear and disputed. What began as a fast-moving global emergency has, by most assessments, settled into an endemic disease pattern, though public health bodies have been cautious about formally declaring that transition complete.
13:00
OncogeneResearchers identified src in a chicken tumour virus as the first confirmed oncogene in 1970, initially assuming it was foreign viral material. In 1976 Dominique Stéhelin, J. Michael Bishop and Harold Varmus found that normal cells across species already carried a closely related gene, a proto-oncogene, showing the virus had captured and altered an ordinary cellular gene rather than introducing something new, work that earned Bishop and Varmus the 1989 Nobel Prize. Robert Weinberg and Mariano Barbacid's 1982 identification of a mutated HRAS gene in human bladder cancer confirmed the same mechanism operates in people, feeding into the multi-hit model in which cancer typically requires several accumulated genetic changes, not a single mutated gene, to become malignant.
8:53
Rosalyn Sussman Yalow1921–2011Radioimmunoassay was a new way to trace biological molecules using radioisotopes. It measured insulin and other substances in blood at previously undetectable levels. That enabled diagnosis of type 1 diabetes, hepatitis screening of blood donors, identification of hormone-related health problems, and assessment of drug and antibiotic dosing. It did not claim to explain disease mechanisms or replace all existing assays. Its power lay in quantitative precision — not theory, not cure, but detection.
13:00
History of penicillinFrom 1939, Howard Florey assembled a team at Oxford, including Ernst Chain and Norman Heatley, that purified enough penicillin to cure infected mice in May 1940 and, in February 1941, to treat a human patient, Albert Alexander, whose initial recovery ended in death only because supplies ran out. Florey and Heatley then travelled to the United States in 1941 to find manufacturers, where a Peoria laboratory's deep-tank fermentation method and a higher-yielding mould strain found on a market cantaloupe let American production scale from a research curiosity to billions of units a year by 1944. Chain wanted to patent the production process to protect British interests, but Florey and British colleagues considered profiting from the discovery unethical, so American firms patented their methods instead, later forcing British manufacturers to pay them royalties, a decision whose costs became clear only once the antibiotic era Florey's team had done so much to launch was already under way.
Up next in Science

Inert-pair effect

· 13:00

Thallium would rather give up one electron than the three its position on the periodic table suggests it should. Sidgwick named the pattern in 1927; nobody has ever reduced it to a single cause.

Play nextBack to Medicine
13:00