Enhancing individual-level consistency of breast cancer Polygenic Risk Score (PRS) through machine learning approaches – GERA

PhD Recipient: Di Mu, Melbourne School of Population and Global Health, University of Melbourne

Supervisors: Associate Professor, Shuai Li, Dr James Dowty, Professor Mark Jenkins, Dr Sibel Saya, Dr Jiadong Mao

Awarded: $10,000 in GERA’s Grant Round 1 (2025)

About the project: Polygenic Risk Scores (PRS) are increasingly used to predict breast cancer risk. While many PRSs perform well at the population level, they can sometimes give different risk estimates for the same individual. This inconsistency presents an important challenge for clinical use. Di’s research will explore how machine learning approaches can improve the consistency and reliability of PRS at the individual level. The project will also investigate how personal characteristics, such as age, ancestry and clinical profile, may interact with PRS models in complex ways. Improving consistency is essential for building trust in genetic risk prediction and supporting the safe integration of PRS into clinical and public health practice.

Go back to:

All Research Projects, Research Projects on Cancer, Research Projects on Methods and Programs, Research Projects on Women’s Health