By combining laboratory biology with machine learning, a 17-year-old student from Virginia has gained new insights into how protein production can go wrong in cancer cells. Aashritha Penumudi, a senior at Thomas Jefferson High School for Science and Technology in Alexandria, studied a process called ribosome stalling. Her research earned her a place among the 40 national finalists in the 2026 Regeneron Science Talent Search, one of America’s leading science and mathematics competitions for high school seniors. The project, titled “Understanding the Structural Basis of Ribosome Stalling by Cellular Arresting Peptides,” examined how ribosomes slow down while making proteins. This process helps control polyamines, small molecules that are important for normal cell growth but can also support aggressive cancer growth when their levels become too high.
Creating stalled ribosomes in the lab
Ribosomes act as tiny factories inside cells. They read genetic instructions and build proteins one amino acid at a time. When certain sequences, known as cellular arresting peptides, move through the ribosome, the machinery can pause or stop. To study this process, Penumudi created stalled ribosomes in the laboratory. Working with her research mentor, she used high-resolution microscopy to take detailed images of the paused ribosome structures. The images showed that certain arrangements of amino acids can physically cause the ribosome to stop as it moves along the messenger RNA molecule.
Using AI to study breast cancer data
Penumudi then used machine learning to test whether her laboratory findings could also be seen in larger sets of cancer data. She trained an artificial intelligence model using publicly available genetic and molecular data from human breast cancer cells. The model analysed peptide sequences and predicted which amino acids were most likely to cause ribosomes to stall in tumour environments. Penumudi compared the AI predictions with the ribosome structures she had observed under the microscope. The computer predictions matched the findings from her laboratory experiments. According to her official finalist profile published by the Society for Science, the combined laboratory and computer results point to possible new ways of controlling polyamine levels and potentially slowing cancer growth.
Recognition and leadership in Herndon
Penumudi, who lives in Herndon, Virginia, was selected from more than 2,600 applicants across the United States. She is part of a finalist group representing 36 schools in 19 states. Each finalist receives at least 25,000 dollars from the Society for Science and Regeneron. Her work in biochemistry is only one part of her academic and community activities. Penumudi is president of her school’s Technology Student Association and helped lead the chapter to become the top-performing group in Virginia. She also leads the school’s neuroscience and biology clubs. Outside school, she works as a certified Emergency Medical Technician (EMT) with the Warrenton Volunteer Fire Company. She also volunteers with the Centreville Immigration Forum, where she helps organise free health screenings and co-leads conversational English classes for local community members.
Possible implications for cancer treatment
Scientists have long faced challenges in controlling polyamine production because these molecules are needed for normal body functions as well as cancer growth. Completely stopping their production can cause serious side effects. Penumudi’s research offers a more targeted idea. By identifying the exact structural signals that make ribosomes stall during protein production, her work could help researchers find ways to adjust polyamine levels rather than stopping their production completely. Combining AI predictions with structural biology could also make it easier to identify promising peptide sequences before researchers spend time and money testing them in the laboratory. The approach shows how machine learning and laboratory biology can work together to better understand the molecular processes involved in cancer and potentially guide the development of more precise treatments.
