In the final months of 2019, the world had no idea it was about to be blindsided. As SARS-CoV-2 began its silent spread through Wuhan, the global scientific community was operating nearly blind when it came to understanding which viruses were circulating in wildlife and which among them carried the molecular keys to unlock human cells. The surveillance infrastructure simply did not exist.
Now, a landmark study published in Nature Communications reveals exactly the kind of detection system that was missing. By systematically sampling 1,462 bats over three years in Cambodia’s Steung Treng province, an international research team has produced 33 complete sarbecovirus genomes and identified four distinct novel viral lineages, two of which carry spike proteins that can already bind to the human ACE2 receptor. The study, funded by the BCOMING project under Horizon Europe, the French ANR, HERA DURABLE, and the NIH PICREID program, serves as both a scientific breakthrough and a sobering reminder of how close the next pandemic could be.
“It is precisely this kind of systematic, longitudinal surveillance at high-risk animal-human interfaces that the world needed before 2019,” the authors write, and the data backs them up. The DOI reference is 10.1038/s41467-026-75954-1.
The Steung Treng Viral Menagerie
Steung Treng province, a remote region in northeastern Cambodia where the Mekong River meets dense tropical forest, is home to extraordinary bat biodiversity. The study focused on horseshoe bats of the genus Rhinolophus, the same group widely recognized as the ancestral reservoir for the SARS-CoV-2 lineage. Over three years, the team trapped and sampled bats, applying both metatranscriptomic sequencing, an unbiased readout of all RNA present in a sample, and targeted amplicon sequencing to capture complete viral genomes with high fidelity.
The work paid off handsomely. The team reconstructed 33 complete sarbecovirus genomes that cluster into four distinct phylogenetic groups, each associated with a different Rhinolophus bat species. This is not a snapshot; it is a living portrait of viral evolution unfolding in real time. The data show that these lineages are not static. They are migrating and recombining over remarkably short distances and timescales, reshuffling their genetic material in ways that could rapidly produce new variant forms.
This finding carries immediate practical importance. Recombination between co-circulating coronaviruses has been a driving force in the emergence of both SARS-CoV-1 and SARS-CoV-2. Watching it happen in Cambodian bats means scientists can see the raw ingredients of pandemic emergence being mixed together in nature.
The ACE2 Key
Two of the four novel viral groups stood out immediately. Their receptor-binding domains (RBDs), the part of the spike protein that physically contacts host cells, showed high structural similarity to the RBD of SARS-CoV-2 itself. That alone was enough to raise concern, but the team went a step further.
They constructed pseudoviruses: non-replicating viral particles displaying the spike proteins from these novel bat sarbecoviruses on their surface. When they exposed these pseudoviruses to cultured human cells expressing the ACE2 receptor, the spikes worked. The pseudoviruses gained entry, demonstrating that these bat coronaviruses have the molecular capability to infect human cells right now, without requiring any intermediate evolutionary steps.
The pseudovirus assay is the closest thing scientists have to a functional test of zoonotic potential without actually handling live virus. It answers the critical question: can the virus spike actually use the human ACE2 door? For two of the four novel groups, the answer is yes.
This does not mean a pandemic caused by one of these viruses is imminent. Many other factors determine whether a bat virus can spread efficiently among humans, including immune evasion capacity, replication kinetics, and the ability to transmit via respiratory droplets. But the ACE2 binding step is arguably the most important molecular prerequisite, and these Cambodian bat viruses already clear that bar.
A Surveillance Blueprint
What makes the study genuinely paradigm-shifting is not any single finding but the proof of concept it represents. The research team did not stumble upon these viruses by accident. They went to a specific geographic region, identified high-risk bat species, and conducted sustained, systematic sampling over multiple years. They combined metatranscriptomics with amplicon sequencing to maximize genome recovery. They performed functional assays to assess spillover risk. And they used the resulting genomic data to track how these viruses move and evolve through space and time.
This is exactly the kind of pandemic radar system that public health experts have been calling for since COVID-19. But in 2019, comparable surveillance efforts were patchy, underfunded, and rarely connected to functional risk assessment. The contrast is stark.
Before SARS-CoV-2 emerged, virologists knew that horseshoe bats harbored a vast diversity of SARS-related coronaviruses. A 2017 study in China’s Yunnan province had identified viruses in a single cave that contained all the genetic building blocks of SARS-CoV-2. But nobody was running large-scale, multi-year surveillance programs in Southeast Asia. Nobody was building pseudoviruses to test whether novel bat RBDs could bind human ACE2. Nobody was integrating genomic epidemiology with functional validation and spatial tracking. The science existed in scattered corners, but the infrastructure to connect the dots did not.
What Comes Next
The Cambodian bat study makes a powerful case that this infrastructure can now be built. The authors call explicitly for continuous surveillance at high-risk animal-human interfaces. The BCOMING project, under which this work was conducted, is designed precisely to build such a framework, linking pathogen discovery in biodiversity hotspots with pandemic preparedness planning.
But building the radar is only the first step. Keeping it running requires sustained international funding, local capacity building, and political will. The study was supported by an alphabet soup of European and American funding agencies: BCOMING (Horizon Europe grant 101059483), RhinoKHoV, ZooCoV (ANR-20-COVI-000), HERA DURABLE, and the NIH PICREID program (U01AI151758). These collaborations demonstrate what is possible when research networks span continents, but they also highlight the fragility of relying on short-term project funding for what should be a permanent surveillance capability.
For the people of Steung Treng and for the global community, the stakes could hardly be higher. The bats of northeastern Cambodia are not harboring some distant, theoretical threat. They are carrying viruses that have already passed the most critical molecular test for human infection. The genomic and functional data are now in hand. The question is whether the world will act on them differently than it did on the earlier warnings that went unheeded before 2019.
The study published today in Nature Communications (DOI: 10.1038/s41467-026-75954-1) is open access under a CC BY 4.0 license, free for anyone to read, share, and build upon. That commitment to transparency is itself a model for how pandemic science should work. The data is public. The methods are reproducible. The viruses are mapped. The rest is up to us.
Note: Based on reporting from Nature Communications (June 2026) and materials from the BCOMING project.