Most public health models assume diseases spread in a vacuum. They track one pathogen, map its trajectory, and hope for the best. Reality is messier. When two infections collide, they interact, compete, and amplify each other inside human populations in ways traditional single-focus math completely misses.
Enter Alyssa Yu, a 17-year-old senior at Poolesville High School in Maryland. She built a computational framework designed to solve this exact forecasting blind spot. Her award-winning project for the Regeneron Science Talent Search tackles the chaos of simultaneous outbreaks head-on.
The Core Problem with Single-Pathogen Models
Epidemiologists often face co-epidemics, or "twindemics," like seasonal influenza sharing space with COVID-19, or secondary infections hitching a ride alongside HIV. Traditional models fail because they treat pathogens as independent variables.
They aren't. One infection alters the host environment, making it easier or harder for a second pathogen to take hold. If you rely on old forecasting tools during a dual outbreak, your data is fundamentally broken. You miss the compounding transmission rates that happen when populations move through connected transport hubs and cities.
Yu recognized that standard epidemiological tracking tools needed an upgrade. She didn't just tweak an existing spreadsheet. She built something entirely different from scratch.
How Reaction-Diffusion Dynamics Map Two Pathogens
To solve how two distinct diseases move through space and time, Yu turned to an unexpected place: chemical engineering.
Her project, titled When Epidemics Meet: Understanding Spatiotemporal Interactions of Two Pathogens on Metapopulation Networks via Reaction-Diffusion Dynamics, borrows mathematical equations typically reserved for studying chemical reactions.
She used metapopulation networks to break large geographic areas down into interconnected nodes—like towns, transit centers, and cities linked by human travel. As people commute and move between these nodes, the model tracks how the two pathogens spread across geography while simultaneously interacting inside the human host.
Working virtually through the MIT PRIMES-USA program with academic mentor Laura P. Schaposnik, Yu formulated equations that highlight hidden geographic hot spots. These are areas where pathogen interaction causes infection spikes far higher than standard linear math would ever predict.
Smarter Vaccine Distribution When Resources Are Tight
Identifying danger zones is only half the battle. Public health departments never have infinite medical supplies. When an outbreak hits, distributing vaccines evenly across an entire region sounds fair, but math proves it is deeply inefficient.
Yu ran simulations testing different immunization strategies within her network model. Her findings point to a sharper reality. Targeted interventions focused on specific demographic groups and high-traffic transport hubs suppress dual outbreaks faster while using significantly fewer total vaccine doses.
By pinpointing where pathogen synergy peaks, health officials can deploy resources precisely where they stop the chain reaction. That shifts public health strategy from reactive scrambling to proactive geometry.
Why This Student Breakthrough Matters Now
Yu's work earned her a spot as one of the top 40 national finalists in the 2026 Regeneron Science Talent Search, selected from a pool of roughly 2,600 high school applicants nationwide. But beyond the competition accolades, her research highlights a shift in how the next generation approaches global challenges.
Complex global travel, denser urban populations, and climate shifts mean overlapping outbreaks will happen again. Relying on outdated, single-pathogen software is no longer an option.
If public health agencies want to stay ahead of the next major public health crisis, they should look closely at how advanced mathematical frameworks combine network epidemiology with chemical reaction mechanics. Real-world crisis response requires tools built for complexity, not convenience. Stop treating outbreaks like isolated events.
