Students Use CyVerse to Help Predict Mosquito-Borne Diseases

A UA course employed the CyVerse computational resource to develop the proof of concept for a map that predicts the abundance of mosquitoes transmitting diseases such as Zika virus and Dengue fever across the U.S.

By Shelley Littin, CyVerse, Sept. 13, 2017

Students in a University of Arizona cyberinfrastructure course developed a map that predicts the abundance of mosquitoes that transmit diseases such as Zika virus and Dengue fever across the United States.

CyVerse, the National Science Foundation‘s UA-led computational resource for biological sciences, served as the primary data infrastructure platform.

The course, called Applied Concepts in Cyberinfrastructure, or ACIC, is taught every fall semester at the UA by CyVerse co-principal investigators Eric Lyons, an assistant professor of plant sciences in the College of Agriculture and Life Sciences, and Nirav Merchant, director of biocomputing at Arizona Research Laboratories. Both Lyons and Merchant are members of the UA’s BIO5 Institute.

“The goal is to bring together a diverse group of students in order to help a professor, or a team of professors, solve a real-world data challenge,” Lyons said. “What really sets this course apart is that these students are working on a real-world problem, and there is no known solution ahead of time. Every year, I go into it wondering if we’re going to make it out the other side with a successful project.”

The class in the fall of 2016 turned out to be a major success.

The students of ACIC worked with three UA faculty members to take a computational model developed at the UA and incorporate layers of data that describe the biology and life cycle of Aedes aegypti, the mosquito species that carries Zika virus and Dengue fever, among other diseases.

“What the students had to do was download meteorological data, run the model for each pixel on the map corresponding to one location, and then combine results of the model and correlate those based on location to create a new layer for the map,” said class client Joceline Lega, a UA professor of mathematics. “It’s a lot of data.”

Throughout the semester, the students, who came from all majors and backgrounds, used their varied skills to work together to solve the data challenge using CyVerse computational resources.

They have succeeded in laying the ground work for creation of a novel, interactive map that could help researchers and public health specialists understand and predict mosquito abundance — and hence instances of mosquito-borne diseases — across the United States.

“The product that’s being developed by this class will give us predictions of mosquito abundance and allow us to see how those are changing over time, and that in turn informs disease risk with respect to mosquito-borne diseases,” said class client Heidi Brown, an assistant professor in the Mel and Enid Zuckerman College of Public Health and co-principal investigator of CLIMAS, the Climate Assessment for the Southwest. Mike Crimmins, an associate professor of soil, water and environmental sciences in CALS and also a CLIMAS co-principal investigator, was the third client for ACIC.

The students’ work will be extremely helpful for the public, Brown said, by enabling public health workers to remind people when to engage in activities around the home to discourage mosquitoes, such as removing standing water.

The students got something out of it, as well.

“The reason I took this course is because in my research, I have a lot of big data, but I don’t really have the computational facility to extract meaning from all of that data,” said Alison Comrie, who was a UA undergraduate and senior studying neuroscience at the time. “I wanted to experience that for myself.

“Also, as a scientist in the 21st century, I think it would be really hard to just be a biologist and not be able to work in interdisciplinary teams that will really make our results more powerful.”

Said Lyons: “One of the great outcomes of this course are complete software systems, such as this map, that scientists anywhere in the world can now use. Through CyVerse, researchers can now get access to high-performance and distributed computing to analyze their data, which they may not be able to do otherwise.

“Of course, the really fun part is challenging students to do something that nobody has ever done before and watching how they come together to make this happen.”

Posted in AZBio News.