Sunday , April 11 2021

Approach to machine learning to monitor the movement of yellow fever mosquitoes through the environment



Yellow fever mosquito (Aedes aegypti) is a major vector of deadly diseases such as dengue fever, chikungunya and Zika virus, which result in hundreds of thousands of deaths worldwide each year. Because Ae. aegypti prefers to bite humans and there are no vaccines for many of these diseases they transmit, developing methods to control these insects is imperative in the fight against the disease.

In a study recently published in Proceedings of the National Academy of Sciences, a research team led by Yale has developed a new method to track how Ae. aegypti move around. Combining mosquito genetic data and satellite environmental data, the authors mapped “landscape connectivity” – defined as a way in which the landscape facilitates the movement of organisms and their genes over large areas. Researchers have specifically developed a new workflow that more effectively models how Ae. aegypti moving through the landscape of the southern United States

“Connectivity maps allow managers to make informed decisions based on how mosquitoes are likely to move through the landscape,” says Evlyn Pless, a postdoctoral researcher at the University of California, Davis and a doctoral student in Yale’s Department of Ecology and Evolutionary Biology. “Our results suggest that in the southern United States, Ae. aegypti it travels through a combination of natural and human-assisted expansion, exploiting areas that are warmer and flatter, as well as human transport networks. “

Plessco is the author of an article with Giuseppe Amatulli, a scientific researcher at the Center for Research Computing and the Yale School of Environment; Norah Saarman, assistant professor of biology at Utah State University; and Jeffrey Powell and Adalgisa Caccone of the Department of Ecology and Evolutionary Biology at Yale.

It is now the most common method for controlling invasive species such as diseases Ae. aegypti is the use of pesticides that are not environmentally friendly. “We now know that some pesticides harm the environment, including people,” says Saarman. “At the same time, mosquitoes are developing resistance to pesticides that we have found to be safe for the environment.

“This creates a challenge that can only be solved by additional information about where mosquitoes live and how they move.”

One of the most modern methods of controlling the release of genetically modified mosquitoes into existing populations, in an effort to stop the reproduction and spread of the disease. The authors say they expect connection maps like the ones they created will be useful in designing more strategic editions of modified mosquitoes.

“By integrating machine learning with the optimization process, our approach goes beyond the limitations of previous methods and should be useful for more precise planning of vector control actions,” says Amatulli.

The authors also believe that this new advancement could have a wider application, including in the preservation and protection of the environment.

“Connectivity maps can also be key to protecting endangered native species,” says Pless, “for example, in designing corridors to connect fragmented populations.”

Source:

Yale School of Environment

Journal reference:

Pless, E., et al. (2021.) An approach to machine learning for mapping the landscape association in Aedes aegypti with genetic and environmental data. PNAS. doi.org/10.1073/pnas.2003201118.


Source link