Washington State University researchers have developed an AI model that predicts potential animal reservoirs for deadly viruses, potentially preventing future pandemics. The innovative model focuses on orthopoxviruses, highlighting global hotspots and aiming to improve surveillance and proactive measures.
In a groundbreaking achievement, a team of researchers at Washington State University (WSU) has developed a cutting-edge artificial intelligence tool capable of predicting animal species that may harbor and spread viruses that infect humans. This innovation holds potential to revolutionize the prevention and control of future pandemics.
The machine learning model, detailed in a study published in the journal Communications Biology, leverages host characteristics and virus genetics to pinpoint potential animal reservoirs and geographic regions prone to outbreaks. Initially applied to orthopoxviruses — including those causing smallpox and mpox — the model is designed for adaptability to other viruses, enhancing global health security.
“Nearly three-quarters of emerging viruses that infect humans come from animals,” co-corresponding author Stephanie Seifert, an assistant professor in WSU’s College of Veterinary Medicine’s Paul G. Allen School for Global Health, said in a news release. “If we can better predict which species pose the greatest risk, we can take proactive measures to prevent pandemics.”
The study, which identifies Southeast Asia, equatorial Africa and the Amazon as high-risk areas, highlights regions with high densities of potential hosts and low smallpox vaccination rates. Despite the eradication of smallpox in 1980 ending global vaccination efforts, the vaccine’s cross-protection against other orthopoxviruses emphasizes the necessity of targeted immunization strategies to mitigate outbreak risks.
Collaborating on this pioneering project, veterinary medicine graduate student Katie Tseng expressed the model’s versatility.
“While we used the model specifically for orthopoxviruses, we can also go in a lot of different directions and start fine-tuning this model for other viruses,” Tseng, the study’s first author, said in the news release.
The model outperformed previous methodologies solely based on the ecological traits of animals. By incorporating viral genetics, it paints a more comprehensive picture of virus-host dynamics.
“Previous models were more based on the characteristics of the host, but we wanted to add the other side of the story, the characteristics of the viruses,” added co-corresponding author Pilar Fernandez, an assistant professor and disease ecologist at WSU’s Allen School. “Our model improves the accuracy of host predictions and provides a clearer picture of how viruses may spread across species.”
Following the global spread of mpox in 2022, concerns have mounted about orthopoxviruses establishing new endemic areas through new animal reservoirs. Traditional methods of identifying reservoirs via field sampling are resource-intensive and impractical on a large scale. The AI model developed at WSU simplifies this task, directing more efficient and effective surveillance efforts.
“If you are looking for the reservoir for mpox virus in Central Africa, that’s one of the most biodiverse places on Earth, so where do you start?” Seifert added. “If we can use these machine learning models to help us prioritize sampling efforts, then that’s going to be really beneficial in identifying where these viruses are coming from and in understanding the risks they pose.”
The project also involved contributions from Heather Koehler, an assistant professor in the School of Molecular Biosciences at WSU, and collaborators from institutions such as the University of Oklahoma, University College London and Yale University.
Source: Washington State University