Projects by Researcher

Scientific Questions Being Studied

This research aims to explore and validate methods for genetic ancestry prediction using the All of Us dataset. The primary questions are: (1) How well do current ancestry prediction methods perform across the diverse populations represented in All of Us? (2) Can we develop improved methods that better capture ancestry information particularly for underrepresented groups and people from multiple ancestries?
This question is important because accurate ancestry predictions are essential for many clinical applications, including Polygenic Risk Scores (PRS). Current methods were developed primarily using European-ancestry data or smaller sample sizes than are available in All of Us, potentially limiting their accuracy.
The exploratory phase will evaluate existing methods' performance and identify areas for improvement, laying groundwork for developing more accurate and inclusive ancestry prediction tools.

Scientific Questions Being Studied

The purpose of this workspace is to familiarize ourselves with exactly how to incorporate All of Us data into our analyses. As this is the first time we are ever using the Researcher Workbench, we would like to explore exactly how to access things like phenotype level data and to use the Data and Cohort browsers.

We intend to be accessing data associated with Cardiac Heart Disease, but without a specific research plan in this workspace. The work here is intended solely for our own educational purposes and will not be published or used in any downstream studies, etc.

Scientific Questions Being Studied

Cardiovascular diseases remain leading causes of mortality worldwide. While traditional risk factors provide valuable insights into disease susceptibility, they fall short in predicting individual disease trajectories. Polygenic Risk Scores have also emerged as promising tools for aggregating genetic information into clinically meaningful metrics. Current research has primarily focused on using PRS for disease onset prediction, with limited exploration of their utility in predicting disease progression. This gap is particularly significant for CAD, where understanding progression could inform therapeutic strategies and resource allocation. However, the relationship between PRS and disease trajectory remains inadequately characterized.
By leveraging AoU, we will evaluate whether PRS can enhance our ability to predict disease progression in CAD patients. We will then explore similar models in other common disease phenotypes.