Abstract: Polygenic risk score (PRS) has demonstrated its great utility in biomedical research by identifying high-risk individuals for different diseases based on their genotypes, facilitating disease monitoring and prevention. However, the broader application of PRS to the general population is hindered by the limited transferability of PRS developed in Europeans to non-European populations. Many statistical methods have been developed to improve PRS prediction accuracy in non-European populations through integrating results from European populations. In this presentation, I will discuss the statistical models underlying these methods, with a focus on those that use genome-wide association study summary statistics, instead of individual-level data, from different populations. The performance of these methods will be demonstrated through their applications to different traits in non-European and admixed populations.
Bio: Dr. Ertefaie's research endeavors revolve around the overarching theme of developing robust and efficient methodologies, aiming to minimize reliance on restrictive modeling assumptions. At the core of his interest is the development of methods that seamlessly integrate machine learning, ensuring data-adaptive approaches while upholding the principles of valid statistical inference. His specific research area includes causal inference, individualized treatment strategies, instrumental variable analyses, high-dimensional data analysis, post-selection inference, and survival analysis.
.png){kind=logo}
