There is an additional weapon in the monitoring of avian influenza: two years after the spread of the H5N1 strain in livestock and sporadic cases recorded in humans, an innovation arrives from Italy destined to strengthen the ability to promptly identify viral evolutions at risk of spillover.
It is called FluWarning and is an early warning system developed by a team from the Polytechnic of Milan and the University of Milan, recently presented on the pages of Science Advances.
New warning system against avian flu
The objective is simple: to recognize in advance those genetic changes that could prelude the jump of species, i.e. the passage of the virus from one animal to another, up to human beings.
The H5N1 virus has been circulating among birds for some time, but in the last two years it has also started to spread among dairy cattle, especially in the United States, where a state of emergency was declared in California in December 2024. Experts’ fear regarding avian influenza is the now well-known spillover, an evolution that would allow the virus to adapt to new hosts and, in the worst case scenario, acquire the ability to transmit from human to human.
The context is intertwined with another phenomenon: in addition to avian flu, we are monitoring the increase in cases of seasonal flu, which in Italy is experiencing an early peak, worsening the epidemiological pressure.
How FluWarning was born
The FluWarning system was created as part of the Prin–Pnrr 2022 program, within the Sensible project, coordinated by Anna Bernasconi (Polytechnic University of Milan, Deib Department). FluWarning exploits data from the Gisaid platform, the large international archive of viral sequences, and uses an advanced statistical approach to learn to recognize the genetic profiles considered “normal” for influenza viruses.
When a new sequence has significant deviations, the software automatically issues an alert that also warns of bird flu. At that point the analysis of virologists takes over, who verify whether the observed deviations could be compatible with a jump in species or with the emergence of a new strain more adaptable to humans.
Bernasconi underlines the flexibility of the tool: it is easily installed, allows personalized geographical and temporal analyzes and allows monitoring updated on a daily basis.
Field trials: from the 2009 swine virus to H5N1 in the USA
FluWarning was initially tested in light of data from the 2009 H1N1 swine flu pandemic, a landmark case of a virus that successfully jumped species. Once tested, the system was applied to the current H5N1 scenario.
According to Matteo Chiara, professor of the Department of Biosciences at the University of Milan, the software identified in advance several clusters of viral activity in various American states, with particular evidence in California. Some alerts were registered before the official reports, reporting specific mutations in the hemagglutinin gene, the protein that allows the virus to attach to cells and infect them. Chiara highlights that FluWarning managed to identify markers characteristic of the Californian strains, helping to outline the evolution of the virus in real time.
By making FluWarning accessible to laboratories and institutions, the aim is to create a surveillance infrastructure capable of intercepting the warning signals of new large-scale viral threats, primarily regarding avian influenza.
Meanwhile, Italian virologists such as Fabrizio Pregliasco and Matteo Bassetti draw attention to the increase in seasonal influenza and the need to strengthen vaccination among the most vulnerable groups.
