Congratulations Top Regeneron Science Talent Search Winners!

Regeneron and Society for Science announced the top ten winners of the Regeneron Science Talent Search, headed by Neel Mougdal, 17, of Saline, Michigan, who won the $250,000 top award.

See the Press Release

Watch the Awards Ceremony

Learn more about the winners’ projects by visiting the Virtual Public Exhibition of Projects!

The Top 3 Award Winners at the 2023 Regeneron Science Talent Search: Neel Moudgal (center)Emily Ocasio (left) and Ellen Xu (right)

The Top Ten 2023 Regeneron STS Winners

First Place: Neel Moudgal

Award Value: $250,000
City, State: Saline, MI


Neel Moudgal, 17, of Saline, developed a computer model that can predict the structure of various RNA molecules to make it easier to diagnose and treat disease for his Regeneron Science Talent Search computational biology and bioinformatics project. Existing tools predicting RNA structure use measurements of the magnetic vibration of atoms known as “chemical shift data” collected from a nuclear magnetic resonance spectrometer. These methods rely on assigning chemical shift values to each atom in a given molecule, which sometimes proves impossible. Neel’s computational model contains a library of possible shapes for a given RNA molecule based on its atomic structure. A statistical method assigns weights to each structure in the library, favoring structures that closely resemble experimental data. By using 2D histograms for reweighting, his model eliminates the need to assign chemical shift data.

Second Place: Emily Ocasio

Award Value: $175,000
City, State: Falls Church, VA


Emily Ocasio, 18, of Falls Church, used artificial intelligence and natural language processing (NLP) to measure the impact of a homicide victim’s demographics on press coverage for her behavioral and social sciences project submitted to the Regeneron Science Talent Search. Emily scoured publicly available FBI homicide records from Massachusetts between 1976 and 1984 and matched victims with their corresponding Boston Globe articles. She then used NLP to assess each article for the amount of humanizing or impersonal language used to describe the victim. Coverage is considered humanizing when it includes personal information such as a victim’s family, occupation or interests. Emily’s results showed that news reports included fewer humanizing details in articles about Black male homicide victims than in those about white males. There were also fewer humanizing details in articles about young Black female victims. Emily hopes her findings will help raise awareness of media coverage bias.

Third Place: Ellen Xu

Award Value: $150,000
City, State: San Diego, CA


Ellen Xu, 17, of San Diego, developed a convolutional neural network, a subset of machine learning used for image analysis, to help diagnose the disease for her Regeneron Science Talent Search medicine and health project. As the leading cause of acquired heart disease and coronary artery aneurysms in children between the ages of 1 and 5, a KD diagnosis is based on five visual signs that can easily be confused with other diseases. To improve identification, Ellen pre-trained her model with images of KD and lookalike diseases from the Internet and images provided by parents of KD children. To generate a larger dataset, she altered the images by adding a variety of random photographic transformations. Ellen’s work indicates that her model can distinguish between children with and without clinical manifestations of KD with 85% specificity using a smartphone photo of the child.

Fourth Place: Max Misterka

Award Value: $100,000
City, State: Harrisonburg, VA


Max Misterka, 16, of Harrisonburg, studied q-calculus for his Regeneron Science Talent Search mathematics project. In calculus, students learn about the derivative and some of its rules. In q-calculus, students work with a quantized derivative, called the q-derivative, which has analogous rules. Initially, Max set out to prove properties about a related topic concerning F-de Rham complexes, but had a key insight, and introduced a new tool called the s-derivative. Using this tool, itself a generalization of the q-derivative, Max proved far more than he anticipated by showing that his s-derivative satisfied many of the analogous rules of their counterparts. Max hopes that his s-derivatives will prove useful in quantum physics and spur mathematicians to think of other ways to use and generalize the q-derivative.

Fifth Place: Linden Chi James

Award Value: $90,000
City, State: Durham, NC


Linden Chi James, 17, of Durham, investigated the potential of the thyroid hormone T3 to treat traumatic brain injury (TBI) in humans using wax moth larvae as the model for their Regeneron Science Talent Search cellular and molecular biology project. Wax moth larvae (caterpillars) share physiological similarities with humans, including central nervous system cells and Juvenile Hormone, a caterpillar version of human T3. For their research, Linden inflicted groups of caterpillars, outside the control group, with TBI, treated them with Juvenile Hormone and then analyzed their blood (hemolymph) for concentrations of immune cells and the presence of toxic molecules that are markers of TBI. Agility tests were run before and after treatment to measure and compare caterpillar motor function. They believe their results show that T3 may be a promising treatment for TBI in humans.

Sixth Place: Ambika Grover

Award Value: $80,000
City, State: Riverside, CT


Ambika Grover, 17, of Riverside, engineered a targeted therapy for patients of ischemic stroke, which is caused by blood clots that deprive the brain of oxygen, for the medicine and health project she submitted to the Regeneron Science Talent Search. Currently, tissue plasminogen activator (tPA) is used to treat ischemic stroke. Its shortcomings include the risk of bleeding elsewhere and an inability to stop new clots from forming. To create her targeted therapy, Ambika used magnetic iron oxide nanoparticles coated with a layer of anti-coagulant to prevent more clots from forming and a layer of tPA to break up the clot. She then enclosed the nanoparticles in a shell to create an injectable microbubble and coated it with small proteins, called peptides, which are attracted to blood clots. She observed that her microbubbles were twice as effective as tPA alone at dissolving clots and show promise as a potential ischemic stroke treatment.

Seventh Place: Ethan Zhou

Award Value: $70,000
City, State: Vienna, VA


Ethan Zhou, 18, of Vienna, worked on a type of machine learning for his Regeneron Science Talent Search mathematics project. Ethan’s work is especially useful for algorithms that predict events that are naturally revealed over time. For example, the program may be tasked with predicting Tuesday’s weather and then predicting Wednesday’s weather once it learns whether it was right about Tuesday. Over time, the predictions get better and better as the algorithm learns from its mistakes. Ethan investigated how well these algorithms can perform when confronted with something very unpredictable. He developed new bounds in both the single-variable case, when the prediction depends on a single number like temperature, and multi-variable cases, when a prediction involves more than one numerical input, like temperature, humidity and wind speed.

Eighth Place: Samantha Maya Milewicz

Award Value: $60,000
City, State: Armonk, NY


Samantha Maya Milewicz, 17, of Armonk, studied how the body’s immune reaction to traumatic brain injury (TBI) can lead to secondary injury by damaging the protective blood-brain barrier (BBB) for her Regeneron Science Talent Search neuroscience project. One element of this protection is claudin-5, a protein that helps hold the cells that line the brain’s blood vessels tightly together. For her research, Samantha used a model of the BBB that mimicked the conditions of a TBI. She found that following a TBI, the body overproduces a protein called MMP-9; this causes the claudin-5 to degrade and this decreases the barrier’s protective effect. Using different agents to reduce MMP-9 overproduction, Samantha was able to restore BBB function in the model, providing insight into MMP-9’s potential as a target for future therapeutic development.

Ninth Place: Siddhu Pachipala

Award Value: $50,000
City, State: The Woodlands, TX


Siddhu Pachipala, 18, of The Woodlands, aimed to develop a machine-learning tool that uses patients’ writing to assess their suicide risk and best course of treatment for his Regeneron Science Talent Search behavioral and social sciences project. Treating high-risk patients has been shown to reduce suicide attempts, but existing assessments have led to under-detection. Using linguistic analysis, Siddhu evaluated diary entries from existing suicidality assessment data. He then wrote code to predict someone’s suicide risk based on two language properties. One is syntax – or grammatical qualities – and the other is semantics, or meaning. Of his two models, the one that found correlations between semantics, suicide risk and course of treatment appeared to be more accurate than the one looking at syntax. He believes this suggests that using semantics to gauge psychological health merits future study.

Tenth Place: Thaddaeus Kiker

Award Value: $40,000
City, State: Fullerton, CA


Thaddaeus Kiker, 18, of Fullerton, developed a machine learning approach to predict the presence and properties of quasi-periodic oscillations (QPOs) in black holes for his Regeneron Science Talent Search space science project. While scientists have long known about QPOs and proposed theories for their occurrence, they don’t know exactly why X-ray light from black holes flickers in these ways. Thaddaeus trained his models to make these predictions based on spectral properties like accretion disk temperature. Then he developed open-source software to allow other researchers to apply his methods to their own work. Thaddaeus hopes his work will help “to unlock mysteries about QPOs and their black hole progenitors” when extended to multiple systems simultaneously.