When initial whispers rumbled of a virus afflicting China, followed by the news of an infected nursing home in Seattle, it was easy for us to feel the problem was far away. That is, until early March when cases of COVID-19 mushroomed across America. While the world started locking down, an unlikely group of researchers stepped up. 

Policymakers grappled with questions about locking down and reopening due to the COVID-19 outbreak. Their decisions were guided by simulated models of the spread of the disease though populations and consequent impacts on hospital intensive care unit capacity. However, researchers at the University of Minnesota’s College of Veterinary Medicine (CVM) noticed ways to improve and build on systems policymakers were using and identified a use for their expertise. So, they developed a new application that builds a new angle to COVID-19 modeling approaches.

Powered by refined models

The models originally used for COVID disease spread assumed all people interact randomly, which is not entirely accurate. We interact in predictable ways, and the team’s new modeling approach accounts for that. With the goal of creating a model aimed at answering questions regarding people's interactions, Matt Michalska-Smith, PhD, a postdoctoral associate in the lab of Meggan Craft, PhD, associate professor in the Department of Veterinary Population Medicine at the CVM, started working. 

The team created the Layered Interaction COVID Network Simulator, or LINCS, which models COVID disease spread under specified circumstances. “One of the nice things about making interactions more explicit is you can think about interventions to reduce disease transmission in a more targeted way,” Michalska-Smith says. 

LINCS users can experiment with different parameters and population sizes. The app breaks down into four main tabs. Introduction, Model Description, Sandbox, and Focused Scenarios. Once you have completed the tutorial under Introduction, you can dive into the rest of the application. And exceptionally curious or science-minded users can explore the Model Description details to learn more about “how the sausage is made.”

Within the Sandbox, users can understand the path of COVID outbreaks by experimenting with disease parameters, such as transmission rate, how long people are infectious, and the likelihood of exposed individuals becoming symptomatic. Adjusting parameters in the Sandbox results in a network of interactions within and between households, which predicts the trajectory of disease spread through the population. 

Focus Scenarios allow users to explore how policy decisions and tradeoffs in our own lives affect the dissemination of the disease. For example, policymakers must consider what COVID-19 transmission will look like as we return to office jobs and kids back to school. On this tab, you can layer in different scenarios, adjust the parameters appropriately, and run the simulation, predicting what the resulting disease spread may be.

Disease ecology: a field to watch

“You might think we should send students back to school or people back to work,” Michalska-Smith says. “But, this app can provide officials with some intuition on how their policies might better account for differences in disease spread under these two scenarios. Tools like this app can help policy makers think about how tradeoffs influence where we divert our efforts, and how our choices differentially affect how the disease spreads through the population.” 

The Craft Lab works with disease ecology, often focusing on wildlife and livestock diseases. This field of study assesses how pathogens and hosts interact, particularly with infectious diseases. With a PhD in theoretical network ecology, Michalska-Smith focuses on how network structures and interactions impact the transmission of disease.

Having disease ecologists in the COVID conversation brings valuable, versatile expertise. “Arguably, it doesn’t matter if these nodes in the networks represent swine or humans, as long as the models are based on well-founded assumptions,” says Craft. “Matt’s strong background in network science allowed us to quickly pivot from studying the spread of pathogens in swine transportation networks to creating a custom-made model of pathogen spread in humans interacting in different types of spaces.”

“I was initially hesitant to jump into this project because I felt like I wasn’t someone who studies human diseases,” Michalska-Smith says. “But it’s been nice thinking about what skills or talents I have that I can use to help other people understand this situation better.” 

An example of how disease ecologists model the spread of COVID-19 among populations. This particular model tells the story of how the novel coronavirus might move among people that have gone to a bar, gone home, and then gone into work.  

Early on, the research team hoped LINCS could inform policymakers making decisions about lockdown and reopening procedures. But as the pandemic has progressed and the country has continued reopening, the purpose of LINCS has evolved. Michalska-Smith sees an educational use for the app as well. “We’re now thinking of a general public using it not to necessarily inform their decisions, although of course it can also be used for that, but more as an educational tool. Users are able to think about how the structure of our interactions affects the way the disease spreads.” So, for those of us who remain unclear on how this disease moves and behaves, this tool could help us better grasp what we are up against.

An enduring impact

This tool has far-reaching purposes, too. In the future, the COVID pandemic will likely be taught in future history classes. LINCS could help students understand exactly how the pandemic occurred and evolved, while also learning what policy decisions were made and how those decisions affected the advancement of the disease.     

“With our network model app, the user can visualize household-to-household transmission under different re-opening scenarios,” Craft says. “Decisions about tradeoffs between balancing disease mitigation and socioeconomic well-being can seem overwhelming, and this app allows users to see the disease impacts of those decisions. We’ve created an engaging way to explore the effects of various policy decisions on community spread.” 

LINCS users can adjust transmission rates in various contexts to map out COVID-19's spread, as well as its impact, which is plotted automatically on the graph to the right. The various adjustments for setting is laid out under the Tutorial and Welcome tabs. 

Alongside Michalska-Smith, the LINCS research team includes Marie Gilbertson, DVM, a graduate student in the Craft Lab who has written the text for LINCS, and Lauren White, PhD, an American Association of the Advancement of Science Fellow and Technology Policy Fellow at the United States Agency for International Development Office of HIV/AIDS,* who created the first version of the app. “Both Marie and Lauren have contributed heavily to conceptualizing the app’s structure and motivation,” Michalka-Smith says.

While LINCS is still going through trial phases and receiving early feedback, the team continues working towards a mass-testing release of the app very soon. Considering the fate of the project, Michalska-Smith says, “I would love to have people come and see the app, interact with it, and hopefully learn things. I’d also love to gather their feedback so the app can be improved to make it more accessible and effective at teaching people about disease spread.” 

With the development of a vaccine, various businesses, workplaces, and schools will start re-assessing their protocols. Modeling disease spread under various circumstances will remain critical. And in the event of a future pandemic, this application sets a precedent for tools to help anticipate the characteristics of how a disease can move through populations. No matter how you look at it, the outcome is clear: the rather unlikely team at the Craft Lab has created something with both policy-making and educational capabilities that will have wide-reaching applications for years to come.

*The contents in this article do not necessarily reflect the view of the U.S. President's Emergency Plan for AIDS Relief, the U.S. Agency for International Development, or the U.S. Government.