As these things usually do, it began quietly. A California duck farm. a standard examination. An outcome that no one desired. Health officials confirmed the first cases of H5N9 avian influenza in the United States in late January 2025. Although the announcement didn’t make headlines in most places, virologists were already aware that another flu strain had entered a system that was already dangerously overcrowded.
There is more to H5N9 than just its existence. It’s where it came from. This virus did not appear on its own. As a reassortant, it was basically pieced together from fragments of three distinct influenza strains: H5N1, H7N9, and H9N2. Eight distinct gene segments make up the segmented genomes of influenza viruses, which can swap and recombine when two distinct strains simultaneously infect the same cell. That type of mixing occurred in a live poultry market in Hangzhou, China, in 2013. The genetic equivalent of a parts bin for new viruses, scientists who examined the resulting samples reported discovering several subtypes coexisting inside individual chickens.
This finding, which was published in the Journal of Virology, validated a long-held but seldom documented hypothesis: live bird markets serve as incubators for genetic novelty as well as individual strains. The H5N9 that surfaced had the N9 neuraminidase extracted from the human-infecting H7N9 along with a highly pathogenic H5 gene inherited from H5N1, the same surface protein causing the severe outbreaks that have been sweeping bird populations worldwide. It was a novel combination. It was distinct from any H5N9 strain that had been previously identified, and scientists were open about the fact that they were still unsure of its full potential.
Labs are currently racing to map the genetic fingerprint of every H5N9 sample they can obtain because of this uncertainty. It’s not merely an academic exercise to sequence a new virus’s entire genome. It’s a type of reconnaissance. The importance of mapping the continent-wide spread of H5N1’s dominant D1.1 strain throughout North America was immediately apparent when researchers at St. Jude Children’s Research Hospital did so. When plotted against the migratory paths of infected waterfowl, scattered human cases that had previously appeared to be geographically random suddenly made sense. Epidemiological speculation was never able to make the connections that the data did.
For H5N9, the same reasoning holds true. There is a huge potential for spread because there are about 40 million migratory aquatic waterfowl in North America alone. These birds’ flight paths don’t stop at borders or biosecurity checkpoints, and they can spread the flu in ways that don’t always make them obviously ill. On that California duck farm, H5N9 was discovered alongside H5N1, which is exactly the kind of coexistence that virologists are most concerned about. Reassortment occurs when two strains share a host. That’s how H5N9 initially emerged.
There is some, but not all, comfort in what scientists have discovered thus far. Currently, the H5N9 virus prefers to bind to receptors in bird respiratory tracts rather than human ones; this distinction is crucial for pandemic risk. The virus caused limited mortality at lower doses when the original H5N9 strains were tested in mice. It doesn’t seem to spread effectively among mammals in its current form. However, its highly pathogenic HA cleavage site is structurally identical to some of the most dangerous H5N1 variants in circulation, and it carries a mutation that confers resistance to one class of antiviral medications. This virus has a harsh underlying architecture.
In a more recent study, scientists at EMBL Hamburg described how influenza A viruses physically interact with human cell proteins during active infection—that is, inside intact living cells rather than in test tubes. According to their research, flu viruses are far more clever than previously thot in how they take over cellular machinery. For example, they can dissolve particular nuclear structures to release proteins that the virus can use to replicate itself. They claim that the technique they created may eventually be used with strains like H5N1 that have a higher potential for pandemics, and consequently, viruses like H5N9 that share the genetic heritage of H5N1.
Whether H5N9 will continue to be an animal disease or establish a more comfortable foothold in humans is still unknown. Anyone who tells you otherwise is advancing science, and that question is still unanswered. It is evident that the tools available to scientists today—whole-genome sequencing, in-cell protein mapping, and continental surveillance networks—are more potent than those used during previous bird flu outbreaks. It’s not just a scientific problem. It’s structured. Individual country-level findings are like patches without a quilt in the absence of coordinated international surveillance.
Boundaries are unknown to the virus. No matter which agency is observing, the genome is unaffected. Beneath all the meticulous scientific jargon, there is an uncomfortable reality, which is why scientists are trying to decode H5N9 as quickly as they can.
