Chemist Jixin Chen looked at the rapid spread of COVID-19 early in the pandemic and saw a new opportunity for his kinetics lab, where they study reaction rates.
When he first led the lab in the spring of 2021, undergraduates concluded that social regulations such as lockdowns, face masks, and social distancing were effective ways to slow the rate of spread of COVID. But they also discovered the limitations of the modelling, noting that large numbers of confirmed cases were not necessarily associated with an increasing rate of spread.
Students in the following lab wrote in a journal article regarding their experience that researchers should continue this work when infection and vaccination rates become significant.
And that’s exactly what happened. Students in the Spring 2022 lab extended the mathematical model to make predictions regarding the rate of spread of COVID-19 in the United States with mass vaccination.
They also ran the model for the state of Ohio through the fall of 2022, correctly predicting the surge in cases the state is seeing in late summer.
The second group of lab students also wrote down their lab experiment, this time seeing it published in the Journal of Chemical Education. All the students left the laboratory with the required conditions fulfilled. But they might also add several lines to their resume – for modeling software experience, data analysis skills, and publication in a journal.
In the spring of 2021, when the world moved away, the use of the COVID model allowed Chen’s students to work on their own computers with publicly available data and software.
It worked so well that the undergraduate students submitted a journal article regarding their experience, noting: “The viral spread model is complicated but parameters, such as its reproduction number, Rt, can be estimated with the model. susceptible, infectious or recovered. COVID-19 data for many states and countries is widely available online. This offers students the possibility of remotely analyzing its propagation kinetics.
Chen noted that COVID modeling offered an advantage when it came to explaining the steady-state approximation for some models in the textbook. Students noted that they had benefited from exploring the simulation feature of commonly used Excel software.
“The most surprising and fun thing for me was how accessible research can be. We only used resources and data from free websites, but from there we were able to go deeper and dive into something so relevant to today’s society,” said Emma Lintelman, a rising senior chemistry major with a minor in biological sciences at the College of Arts & Sciences.
In the spring of 2022, Chen and his students took numerical simulation of kinetics and regression modeling even further.
“The first time we used this approach, students were able to apply kinetic techniques learned in physical chemistry to analyze a real-life problem in progress in a distance learning environment,” Chen said. “This year, another group of undergraduate students led by graduate students Dylan Smith and Tharushi Ambagaspitiya did the same practice and extended the mathematical model to predict the spread of COVID-19 in the United States with mass vaccination. . »
In the lab, the Sensitive Infectious Recovery Model (SIR) and the SIR Vaccine Model (SIRV) are explained to students and are used to analyze COVID-19 spread data from the United States Centers for Disease Control and Prevention (CDC). The basic reproduction number R0 and the real-time reproduction number Rt of COVID-19 are extracted by fitting the data to the models, which explains the kinetics of spread and provides a prediction of the trend of spread in a given state.
Students can quickly see the differences between the SIR model and the SIRV model, Chen said. The SIRV model accounts for the effect of vaccination, which helps explain later stages of the ongoing pandemic.
Students also learned the predictive power of models by making predictions for the following months.
“I think the most surprising part of running our COVID-19 kinetics simulation was seeing the drastic effects of variation in reproduction number over time in our simulation,” said David McEwen, a senior majoring in chemistry and minor in business. “This allowed us to directly simulate different levels of virus regulation through masking, social distancing, etc. By dramatically changing the number, we were able to see directly with our data the increase or decrease in the rate of spread of the virus, which was staggering at times.
“I think the hardest parts for me were initially setting up our simulation settings and fitting the simulated data to the collected case data. Fitting simulated data to actual case numbers sometimes required fine-tuning and took some time,” McEwen said.
Lintelman agreed.
“The hardest part for me was fixing the bugs in our formulas,” she said. “It can be tricky when you’ve been staring at your data for hours. It all just starts to swirl around in the mind, but that’s just when you have to come back to it later when you have a clear mind. »
Chen noted that a grant from the National Institutes of Health (Super-Resolution Optical Mapping for DNA Analysis Using Triplex-Forming Oligonucleotides as Stochastic Molecular Probes) partially funded the teaching lab.