Amber Young earned her PhD in Biostatistics at UNC Chapel Hill in 2026, where her research bridged computational statistics, machine learning, clinical trial design, and translational oncology to advance the discovery and targeting of biomarkers in cancer. Working under the co-supervision of Drs. Naim Rashid and Didong Li, she developed semi-supervised learning methods for the discovery of prognostic patient biomarkers and designed biomarker-driven Bayesian adaptive clinical trials. She is now a Biostatistician at Natera.
Amber’s dissertation focused on molecular subtype discovery in pancreatic cancer, creating methods for identifying clinically relevant and replicable cancer subtypes via semi-supervised matrix factorization. Her first-author work on DeSurv, an R package coupling nonnegative matrix factorization with a Cox proportional hazards layer, identifies prognostic gene programs that generalize across pancreatic cancer cohorts (reproducibility materials). Her technical expertise spans clinical study design, C++, machine learning, high-performance computing (HPC), and R programming.
After joining UNC in 2019, Amber was selected as a trainee on the prestigious National Cancer Institute’s (NCI) Cancer Genomics Training Grant program, where she trained for three years (2019-2022) in cutting-edge computational methods for cancer genomics research. This fellowship provided intensive training at the intersection of biostatistics and cancer biology, establishing her foundation for translational research.
Amber gained diverse research experience across multiple collaborations. As a Graduate Research Assistant at UNC Gillings School of Global Public Health (2021-2024), she worked with the Center for AIDS Research (CFAR), applying machine learning and high-performance computing to infectious disease research. Earlier, she served as a Graduate Research Assistant for the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) at the Collaborative Studies Coordinating Center (2019-2020), contributing to large-scale epidemiological research.
In Summer 2022, Amber completed a Statistics and Data Analytics internship at Eli Lilly and Company in Indianapolis, where she developed a model-assisted dose-finding design for pediatric patients incorporating PK/PD (pharmacokinetic/pharmacodynamic) data.
Amber earned her BS in Mathematics and Statistics from Purdue University (2015-2019), where she also served as a Computer Science Teaching Assistant (2018-2019), teaching C programming and supporting undergraduate students in computational methods.
Key Lab Publications
- Young, Amber M., et al. “Differential Transcript Usage Analysis Incorporating Quantification Uncertainty via Compositional Measurement Error Regression Modeling.” Biostatistics, 2024. CompDTUReg (Biostatistics 2024)