Research

Areas of Research

My primary research interests are in Statistical Genetics and Genomics, Risk Prediction, Missing Data, and Semi-Parametric Methods. My thesis research focuses on building better risk prediction models that perform equitably across diverse populations to help promote precision prevention. I currently build and validate cancer risk prediction models using very large and diverse data that combine classical risk factors (e.g. demographics, lifestyle, and environmental variables) and polygenic risk scores in a single model. To support this work, I am developing new transfer learning techniques to facilitate combining information across data sources that exhibit complex patterns of missing data.

Papers in Progress

  1. Norton, E.L., Ahearn, T., Garcia-Closas, M., Kraft, P., and Chatterjee, N. Development of a universal breast cancer risk prediction tool for the diverse US population using data from 2.4 million women across 29 cohorts.
  2. Norton, E.L., Kraft, P., Brantley, K., Ahearn, T., Garcia-Closas, M., and Chatterjee, N. GENMETA-Lite: A Fast and Flexible Approach to Federated Learning with Disparate Covariate Information Across Studies.
  3. Norton, E.L., Ahearn, T., Mukhopadhyay, S., Balasubramanian, J., Kim, E., Li, S., Pal Choudhury, P., Garcia-Closas, M., and Chatterjee, N. Validation and population projection of multi-cancer risk incorporating polygenic risk scores and non-genetic factors.

Recent Presentations

  1. Norton, E.L., Kraft, P., Brantley, K., Ahearn, T., Garcia-Closas, M., and Chatterjee, N. Federated learning with missing covariates and distribution shifts. Joint Statistical Meetings. Boston, MA. August 2026. Invited oral presentation.
  2. Norton, E.L., Ahearn, T., Mukhopadhyay, S., Balasubramanian, J., Kim, E., Li, S., Pal Choudhury, P., Garcia-Closas, M., and Chatterjee, N. Cross-population validation and projection of multi-cancer risk models incorporating genetic and non-genetic factors. American Association for Cancer Research (AACR) Annual Meeting, San Diego, CA. April 2026. Poster presentation: Abstract #6667.
  3. Norton, E.L., Kraft, P., Brantley, K., Ahearn, T., Garcia-Closas, M., and Chatterjee, N. GENMETA-Lite: A Fast and Flexible Approach to Federated Learning with Disparate Covariate Information Across Studies. ENAR 2026 Spring Meeting, Indianapolis, IN. March 2026. Oral presentation.
  4. Norton, E.L., Ahearn, T., Mukopadhyay, S., Balasubramanian, J., Kim, E., Pal Choudhury, P., Garcia-Closas, M., and Chatterjee, N. Modular and interpretable multicancer risk prediction of the 14 most common cancers in the U.S. to identify high-risk individuals missed by current age-based screening guidelines. inHealth Precision Medicine Symposium, Baltimore, MD. May 2025. Poster presentation.
  5. Brantley, K.D., Ahearn T., Norton, E.L., Palmer, J., Zirpoli, G., Neuhouser, M.L., Barnett, M., Teras, L., Hodge, J., Rohan, T., Milne, R., Eliassen, A.H., Huang, H., Chen, Y., O’Brien, K., Kitahara, C., Anderson, G., Lee, I, Chatterjee, N., Garcia-Closas, M., Kraft, P., on behalf of Breast Cancer Risk Prediction Project. Performance of common general-population breast cancer risk prediction models in 15 cohorts. American Association for Cancer Research (AACR) Annual Meeting, Chicago, IL. April 2025. Poster presentation: Abstract #1325.

Awards