Approved Research

Development and validation of polygenic risk score models for Taiwanese population

Lay summary

Polygenic risk scores (PRS) are numerical risk indicators based on multiple genetic markers to provide insight into a patient's inherited risk of developing a disease. PRS is usually constructed using information about multiple genetic variants together with the results from genome-wide association studies (GWAS). Our hypothesis is that interactions between genetic variants could be crucially contributed to complex traits and disease. We have developed a novel method, GenEpi, to consider the paired interactions between genetic variants with machine learning approaches. Here in this project, we will perform the following aims within three years. 1)We intend to use the UK biobank to further improve the applicability of our GenEpi method. 2) We would apply our GenEpi method together with other popular methods to construct PRS in the Taiwanese population for 17 quantitative trait measurements and several diseases. 3) We compare the predictive performance of PRS in UK biobank with that in our Taiwanese dataset to examine the difference between different ethnicity/races. Our project is expected to help further application of PRS by comparing different methods and this will bring a new perspective to interpret the genetic variations. This proposal is consistent with UK Biobank's mission of health-related research in the public interest. Furthermore, we may provide a novel PRS combined with clinical information associated with each individual in collaboration with local hospitals in Taiwan. Therefore, we are interested in all the dataset collected by the Biobank for studying complex traits and diseases. We expect to advance and accelerate the use of PRSs in clinic practice.