Between 2016 and 2018, the schooner Tara sailed more than 100,000 kilometers across the Pacific Ocean, stopping at coral reefs from the Great Barrier Reef to the Panama coast. Scientists aboard collected thousands of coral fragments, froze them, and shipped them to sequencing labs on three continents. What those labs found is reshaping how biologists think about reef life: each coral species harbors a bacterial community so distinct, and so poorly known, that the majority of recovered microbial genomes had no close match in any existing database.
The central study, led by microbiologist Pierre E. Galand and colleagues, was published in May 2026 in Nature under the title associated with the Tara Pacific expedition’s coral microbiome analysis. The team sampled 32 coral species across the Pacific and applied genome-resolved metagenomics to reconstruct thousands of near-complete microbial genomes directly from coral tissue. They assembled these into the Reef Microbiomics Database, or RMD. Of the microbial genomes recovered, roughly three-quarters represented previously undocumented lineages with no close match in public repositories. Across all sampled coral species, the pattern was consistent: bacterial residents of one host differed sharply from those of another, even when the corals grew side by side on the same reef. (Note: the Nature link points to the journal’s homepage because the specific article URL was not available at the time of writing; readers can search the journal site for the Tara Pacific coral microbiome paper for the full text and DOI.)
Corals as microbial curators
The idea that corals do not simply absorb whatever microbes drift past them gained further support from a 2024 study in Nature Communications focused on a single species, Acropora kenti. Researchers reconstructed metagenome-assembled genomes from coral tissue and compared them with genomes from the surrounding seawater. The coral hosted bacterial lineages that were absent from the water column entirely, suggesting the animal actively recruits and maintains specific partners rather than passively filtering the ocean.
Earlier work had hinted at this selectivity. A foundational study archived in the NOAA Institutional Repository (Ainsworth et al., 2015) showed that corals host highly diverse microbial communities while sharing a small set of rare bacterial phylotypes across vast distances. Those shared taxa, sometimes called a “core microbiome,” appear to persist regardless of geography, pointing to a conserved partnership that may be essential for basic coral physiology even as other community members vary from species to species.
A separate database effort, described in a 2018 paper indexed in PubMed, integrated coral-associated ribosomal RNA sequences with host and location metadata. That project demonstrated that microbial communities tracked both biology and geography, not random chance, and it did so at a scale that foreshadowed the genome-level catalogs now emerging from the Tara Pacific data.
The leap from marker-gene surveys to full genome reconstruction matters because it changes the questions scientists can ask. Knowing a bacterial group is present tells you little about what it does. Recovering its genome lets researchers predict metabolic roles, identify biosynthetic gene clusters, and begin to understand how microbial chemistry might influence whether a coral survives a marine heatwave or succumbs to disease.
Promising chemistry, but a long road to the pharmacy
Press materials distributed through EurekAlert described coral reefs as “an endangered natural pharmacy,” and the label is not baseless. The Tara Pacific study identified novel gene cluster families in coral-associated microbes, including clusters that encode enzymes and peptides with potential biomedical relevance. Biosynthetic gene clusters are the molecular assembly lines behind many existing antibiotics and anticancer agents, so finding new ones in an unexplored habitat is genuinely exciting to drug-discovery researchers.
But excitement and evidence are different things. As of June 2026, no specific compound isolated from these coral microbiomes has entered preclinical testing, let alone a clinical trial. The predictions rest on computational analysis of gene content, not on laboratory experiments with purified molecules. The distance between a predicted gene cluster and a usable drug is typically measured in a decade or more of synthesis, cell-based assays, animal testing, and regulatory review. Readers should treat the pharmaceutical angle as a scientifically grounded hypothesis, not a near-term promise.
Open questions scientists are still working through
Several gaps in the evidence deserve attention. The Acropora kenti study is compelling for that single host, but corals span hundreds of genera with wildly different body plans, habitats, and stress tolerances. Massive boulder corals, delicate branching forms, and plate-like species all interact differently with light, currents, and sediment. Whether every coral species maintains the same degree of microbial exclusivity, or whether some are more cosmopolitan in their bacterial partnerships, remains unresolved. Comparable genome-resolved surveys across many more hosts will be needed before anyone can claim universal patterns.
A related puzzle involves reconciling the “core microbiome” concept from earlier NOAA-archived research with the host-specificity emphasis of the genome-resolved work. These findings are not necessarily contradictory. A coral species could share a handful of core partners with distant relatives while also hosting dozens of unique ones. But no published analysis has yet mapped genome-resolved lineages onto the earlier marker-gene phylotypes using the same dataset. Until that reconciliation happens, the field is working with two complementary but unmerged frameworks.
There is also the tantalizing but unproven idea that corals might acquire microbes from neighboring species during environmental stress, enabling a kind of rapid microbial adaptation that static surveys would miss. Demonstrating such exchange would require time-series sampling during heatwaves or pollution events, tracking specific strains as they move between hosts. No study in the current literature has done this for coral-to-coral transfer at the reef scale, so the concept remains speculative.
Why losing a coral means losing more than a skeleton
For conservation, the practical implication is straightforward but sobering. Protecting a coral species does not just preserve an animal and the reef structure it builds. It also safeguards an entire microbial consortium, many members of which appear to exist nowhere else. When a coral colony dies from bleaching, disease, or coastal pollution, its microbiome likely vanishes with it, erasing genetic material that may never reappear elsewhere in nature.
The loss is compounding. Reef ecosystems lose the engineering services that living corals provide: fish habitat, wave buffering, nutrient cycling. And the broader scientific community loses access to a vast, largely unexplored library of microbial genes and chemistries whose functions are only beginning to come into focus. As genome-resolved surveys expand to more species and more oceans, they will sharpen not only our understanding of coral biology but also the stakes of allowing reefs, and the invisible life they shelter, to disappear before we know what we are losing.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
