A new analysis published in FEMS Microbiology Letters has found that more than 70% of Earth’s ecoregions have no publicly available soil sequencing data for arbuscular mycorrhizal fungi, the underground networks that help most plants absorb nutrients and store carbon. The finding, drawn from the largest global environmental DNA dataset for these organisms, means that vast stretches of the planet’s drylands, steppes, and grasslands remain biological blank spots. For a kingdom of life estimated to contain between one and five million species, the gap raises hard questions about what science is missing and what conservation efforts are failing to protect.
A Kingdom Barely Cataloged
The Fungi kingdom has been estimated to include between one and five million species, yet only around 5% are formally described. That number alone signals how little researchers know about the organisms that quietly sustain terrestrial life. Roughly 90% of plants depend on fungal partnerships to thrive, and many human medicines, including penicillin, trace their origins to fungi. The structure, health, and existence of terrestrial ecosystems depend on these organisms in ways that science is only beginning to measure.
The new study in FEMS Microbiology Letters zeroed in on arbuscular mycorrhizal (AM) fungi, a group that forms intimate associations with plant roots and helps shuttle phosphorus and other nutrients from soil into vegetation. Using the largest global AM-fungal eDNA compilation, the researchers mapped which of Earth’s ecoregions have been sampled with modern DNA metabarcoding and which have not. The result: only about 30% of ecosystems have any publicly available AM-fungal sequencing data at all. Drylands, including deserts, steppe regions, and grasslands, are among the most glaring blind spots, according to coverage of the findings, leaving some of the world’s most climate‑sensitive landscapes effectively invisible to fungal science.
The Data Behind the Gap
The backbone of this analysis is the GlobalAMFungi database, an open-access resource that compiles roughly 8,500 samples and about 50 million observations drawn from around 100 original studies. The database uses multiple barcode regions, including SSU, ITS2, and LSU, to detect AM fungi in soil, allowing researchers to capture both broad taxonomic patterns and fine‑scale genetic variation. Its sibling project, GlobalFungi, is even larger, containing more than 600 million observations from over 17,000 samples across 178 studies worldwide. Together, these datasets represent the most extensive attempt to catalog fungal life from environmental DNA and to put underground biodiversity on the same quantitative footing as better‑known plant and animal surveys.
Yet even at that scale, the geographic coverage is heavily skewed toward well-funded research regions in Europe and North America, leaving most of the tropics, boreal zones, and arid landscapes essentially uncharted. The FEMS Microbiology Letters analysis shows that entire ecoregions in Africa, Central Asia, and parts of South America have no AM-fungal sequencing data deposited in public repositories. That bias reflects where research money, sequencing facilities, and long-term ecological plots are concentrated. It also means that global models of soil fertility, carbon storage, and climate feedbacks are being built on a patchwork that largely omits the very regions expected to experience the fastest environmental change.
Invisible Diversity and “Dark Taxa”
The problem is not just missing maps. Even in regions where sequencing has been done, a large share of the fungal diversity detected does not match anything in existing reference libraries. A study in Current Biology documented the prevalence of so‑called dark taxa in ectomycorrhizal fungi—species that appear as genetic clusters in environmental datasets but cannot be linked to any named organism. The authors showed that many of these unclassified lineages are widespread and ecologically important, yet they are effectively absent from conservation assessments because they lack formal taxonomic identities. In practice, that means even the sampled 30% of ecoregions contains a hidden majority of fungal life that science has not yet cataloged.
For AM fungi, the same pattern almost certainly holds. Environmental DNA surveys routinely recover sequences that fall outside curated databases, hinting at a vast reservoir of undescribed species. Without physical specimens, cultures, or detailed ecological data, those lineages remain statistical ghosts, present in models but missing from field guides, red lists, and policy frameworks. The combination of geographic under‑sampling and taxonomic darkness creates a double blind spot: many ecosystems have never been surveyed, and even where they have, much of the diversity detected cannot be named, monitored, or legally protected.
Why Underground Networks Shape Life Above Ground
The stakes of these blind spots extend well beyond taxonomy. AM fungi form dense networks that connect plant roots across entire ecosystems, shuttling nutrients, water, and chemical signals between individual plants. Western University biology professor Greg Thorn has put it bluntly: “A lot of ecologists are beginning to realize that mushrooms really run the world. We’ve quite naturally spent a lot of our time focusing on plants and animals and birds.” That shift in attention is driven by accumulating evidence that fungal networks regulate plant community composition, soil carbon storage, and ecosystem resilience to drought and disease, functions that become more critical as climate extremes intensify.
A 2025 study in Nature used machine learning trained on roughly 25,000 geolocated soil samples and more than 2.8 billion fungal DNA sequences, compiled from both GlobalFungi and GlobalAMFungi, to map global hotspots of mycorrhizal richness. The results showed that many of the richest predicted zones fall outside protected areas, prompting researchers to call for urgent conservation action. Reporting by one major newspaper underscored the policy gap: biodiversity frameworks tend to focus on charismatic plants and animals, while the organisms that keep those species alive receive little formal recognition. If conservation planning continues to ignore fungal networks, efforts to safeguard forests, grasslands, and croplands may miss the very infrastructure that underpins their stability.
Unsampled Regions Could Hold Medical and Agricultural Breakthroughs
The consequences of ignoring unsampled ecosystems are not limited to conservation. Fungi have long been a source of antibiotics and other bioactive compounds, with classic examples such as penicillin emerging from relatively routine laboratory work. A recent overview of natural products research notes that fungal metabolites remain a rich source of candidate molecules for new antibiotics, anticancer agents, and immunosuppressants. Many of those compounds come from species adapted to extreme or unusual environments (exactly the kinds of drylands and steppes that the AM-fungi analysis identifies as unsampled). If those regions harbor unique symbiotic fungi, they may also hold untapped chemical diversity with direct medical relevance.
Agriculture stands to gain as well. AM fungi are already being explored as biofertilizers that could reduce dependence on synthetic phosphorus and nitrogen inputs, improving yields while cutting greenhouse gas emissions from fertilizer production. However, most commercial inoculants are based on a narrow slice of well-studied species, often isolated from temperate croplands. Without broader sampling, researchers cannot know whether more efficient, stress‑tolerant, or crop‑specific partners exist in other ecoregions. In a warming world where farmers are grappling with drought, salinity, and degraded soils, failing to explore the full global repertoire of mycorrhizal symbionts may mean leaving powerful adaptation tools undiscovered.
Closing the Underground Data Divide
Addressing these gaps will require more than isolated field campaigns. The FEMS Microbiology Letters team argues that coordinated sampling strategies, tied to standardized protocols and open data requirements, are essential to make new surveys comparable with existing databases such as GlobalAMFungi. That means funding agencies and research institutions must prioritize long‑term collaborations with scientists in under‑sampled regions, ensuring that sequencing capacity and data ownership are shared rather than centralized in a handful of wealthy countries. It also means integrating fungal sampling into broader biodiversity and soil monitoring programs, so that every new plot or observatory automatically contributes to closing the underground data divide.
Public engagement and policy shifts will matter too. Fungi rarely feature in environmental education, citizen‑science projects, or mainstream conservation campaigns, making it harder to build political support for their protection. Yet the same media outlets that have elevated fungal conservation in recent years also depend on reader backing to sustain in‑depth science coverage; initiatives that encourage audiences to support independent reporting and to engage more deeply with coverage help keep these issues in the public eye. Ultimately, recognizing fungi as foundational to food security, climate stability, and human health may be the most effective way to ensure that the unseen majority of Earth’s ecosystems is finally brought into scientific view.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
