A team of researchers has developed a new digital system called FluWarning to monitor the cross-species transmission of avian influenza. This innovative tool is designed to identify abnormal changes in flu viruses, particularly those that may indicate a spillover from birds to other animals or humans. The system was created by a collaborative team from the Politecnico di Milano and the University of Milan and is detailed in a recent study published in the journal Science Advances.
The FluWarning system analyzes the genetic code of influenza viruses to detect subtle changes that could suggest a potential cross-species transmission. This project falls under the SENSIBLE initiative (Small-data Early warNing System for viral pathogens In puBLic hEalth), which aims to enhance public health surveillance. The research team includes Anna Bernasconi, the Principal Investigator, as well as Professor Stefano Ceri and Researcher Tommaso Alfonsi from DEIB—Department of Electronics, Information and Bioengineering at the Politecnico di Milano, alongside Matteo Chiara, a Professor in the Department of Biosciences at the University of Milan.
How FluWarning Functions
The functionality of FluWarning relies on data from the GISAID platform, which provides access to global virus sequences and related metadata. The system was initially developed using data from the H1N1 swine flu pandemic in 2009, a well-documented case of zoonotic transmission, and has since been adapted to monitor H5N1 bird flu. This highly pathogenic strain has begun to affect cattle in the United States, prompting increased vigilance.
FluWarning employs statistical methods to identify anomalies in flu virus sequences. By learning the normal sequences of influenza, it generates alerts when significant changes occur. Depending on its settings, the system can detect either individual anomalous sequences or groups of them. When an alert is triggered, virologists can then examine the sequences to confirm or dismiss the possibility of spillover.
Recent Findings and Impact
In 2024 and 2025, two genotypes of H5N1 were linked to separate outbreaks in the United States, where numerous dairy cattle were found to be infected. The FluWarning system played a crucial role in identifying clusters of viral activity across multiple states, particularly in California. On December 18, 2024, a state of emergency was declared in California due to the heightened risk of bird flu contamination among cattle. Notably, some alerts from FluWarning were issued prior to official reports, highlighting the system’s potential to provide early warnings.
Additionally, FluWarning detected specific mutations in the hemagglutinin (HA) gene, a vital protein influencing how the virus infects host cells. Matteo Chiara emphasized that the tool effectively monitored the virus’s evolution and identified markers characteristic of the strains present in California.
This advancement represents a significant step forward in the detection of viral changes that could threaten both animal and human health. Stefano Ceri stated, “By making this technology widely available, we hope to strengthen global surveillance on a health issue of global significance.” With its user-friendly installation and the ability to analyze data for specific locations and timeframes, FluWarning is poised to aid laboratories and regional genomic surveillance institutions in making impactful discoveries.
The research team believes that FluWarning’s operational capabilities can provide daily feedback on viral changes, enhancing the ability to respond to potential health threats effectively. For more detailed insights, refer to the study by Tommaso Alfonsi et al., titled “Lightweight multiscale early warning system for influenza A spillovers,” published in Science Advances in 2025.
