A red-tailed guenon in Bwindi Impenetrable Forest region in Uganda chews on a...
A red-tailed guenon in Bwindi Impenetrable Forest region in Uganda chews on a rope coated in jam, which is a noninvasive sampling technique.
Source: T. Smiley Evans/UC Davis

AI identifies likely wildlife hosts for emerging flaviviruses

After collecting data and comparing it with every known mammal and bird species on Earth, scientists have identified wildlife species that are the most likely to host flaviviruses such as Zika, West Nile, dengue and yellow fever. Flaviviruses are known to cause major epidemics and widespread illness and death throughout the world.

The resulting “hot spot” maps show regions of the world with high diversity of potential wildlife hosts of flaviviruses — viruses mostly spread by mosquitoes and ticks. These include regions where flaviviruses have not been detected but that have wildlife species with the potential to harbor them.

The information provides scientists and health authorities with a road map for disease detection and surveillance efforts. “Tomorrow, if there’s an outbreak anywhere in the world, we now know which wildlife species are most likely to be infected in addition to humans,” said lead author Pranav Pandit, a postdoctoral scholar with the UC Davis One Health Institute’s EpiCenter for Disease Dynamics in the School of Veterinary Medicine.

This map, Fig. 4 in the study in Nature Communications, shows the geographical...
This map, Fig. 4 in the study in Nature Communications, shows the geographical distribution of predicted flaviviral host richness. A) Yellow fever virus (YFV) and Zika virus (ZIKV). B) West Nile virus (WNV), St. Louis encephalitis virus (SLEV) and Usutu virus (USUV). C) Tickborne encephalitis virus (TBEV). D) Rio Bravo virus (RBV), Entebbe bat virus (ENTV) and Dakar bat virus (DBV). E) Dengue virus (DENV). F) Japanese encephalitis virus (JEV).
Source: Maps were generated using data from International Union for Conservation of Nature, BirdLife International and NatureServe.

Predicting potential hosts

Recently Zika virus emerged and continues to circulate in South America and Southeast Asia. The study predicts potential wildlife hosts in these regions with the ability to maintain Zika virus transmission in nature. There is also rising concern that Japanese encephalitis virus will emerge and establish in Europe. The study identifies Europe as one of the regions with a high richness of potential Japanese encephalitis hosts, including many common bird species.

For the study, researchers collected all the published data on wildlife species that have tested positive for flaviviruses. They identified important host traits, such as environmental and physiological features. Then they used a machine learning model that considered the roughly 10,400 avian and 5,400 mammal species in order to identify the most likely species to host viruses. The model predicted hundreds of previously unobserved host species. For example, it predicted 173 host species for dengue virus, of which 139 have not been previously recognized.

Co-leading author and UC Davis professor Christine Kreuder Johnson said the modeling work can help researchers identify which primate species could be potential virus hosts. For example, the model indicated that primates are the main hosts of Zika and yellow fever, but only nine of the 21 primate species predicted to be hosts have been detected with either of those viruses due to limited surveillance activities among these species to date. UC Davis One Health Institute scientists have established noninvasive sampling techniques for primates, such as collecting saliva from sticks and plants chewed by primates or from ropes coated with strawberry jam. But flaviviruses can be difficult to detect, especially in wildlife. “We needed this modeling technique to help us understand the most likely hosts for these viruses in their natural habitat,” said Johnson, director of the EpiCenter for Disease Dynamics. “That’s important for both global health and wildlife conservation. Many of these primates are already endangered, and these diseases burden an already strained population.”

Subscribe to our newsletter

Related articles

Using machine learning to estimate COVID-19’s seasonal cycle

Using machine learning to estimate COVID-19’s seasonal cycle

Scientists are launching a project to apply machine learning methods to assess the role of climate variables in disease transmission

Acute chikungunya infection studied with AI

Acute chikungunya infection studied with AI

Computational tools applied to biology are revolutionizing the study of what happens inside cells during an infection, helping scientists to understand disease mechanisms.

A smartwatch-based algorithm to detect viral infections

A smartwatch-based algorithm to detect viral infections

Purdue University engineers and physIQ have developed a viral detection algorithm for smartwatches.

Enabling AI-driven advances without sacrificing privacy

Enabling AI-driven advances without sacrificing privacy

Secure AI Labs is expanding access to encrypted health care data to advance AI-driven innovation in the field.

Expanding human-robot collaboration in manufacturing

Expanding human-robot collaboration in manufacturing

To enhance human-robot collaboration, researchers at Loughborough University have trained an AI to detect human intention.

AI finds out what makes human tick

AI finds out what makes human tick

Scientists have developed a machine learning technology to understand how gene expression regulates an organism's circadian clock.

Deep learning predicts viral infections

Deep learning predicts viral infections

Using fluoresence images from live cells, researchers have trained an artificial neural network to reliably recognize cells that are infected by adenoviruses or herpes viruses.

Designing better antibody drugs with machine learning

Designing better antibody drugs with machine learning

Artificial intelligence could help to optimise the development of antibody drugs. This leads to active substances with improved properties, also with regard to tolerability in the body.

Using AI to find new uses for existing medications

Using AI to find new uses for existing medications

Scientists have developed a machine learning method that crunches massive amounts of data to help determine which existing medications could improve outcomes in diseases for which they are not prescribed.

Popular articles

Subscribe to Newsletter