Project Summary
Alzheimer's disease (AD) is a major public health concern in the US, and there is an urgent need for reliable,
inexpensive, and non-invasive biomarkers to identify individuals at risk for AD. This project aims to develop
innovative computational strategies for precision medicine in AD. Specifically, we will harmonize several large
longitudinal clinical datasets, identify DNA methylation (DNAm) biomarkers for cognitive reserve (CR), and build
DNAm-based prediction models for AD. In this project, we will improve the accuracy of DNAm-based prediction
models by leveraging knowledge of cognitive reserve, harmonizing multiple datasets, and training and testing
prediction models using samples from longitudinal studies. The DNAm-based prediction models will provide an
inexpensive and convenient approach for identifying subjects most likely to progress to clinical AD, reducing
heterogeneity in patients selected for clinical trials, and facilitating personalized treatment strategies in AD.
Additionally, identifying DNAm markers for CR will help develop novel therapeutic targets and lifestyle
interventions for preventing dementia. The successful completion of the project will also provide the community
with harmonized datasets for AD research, as well as computational methods and tools that can be easily
adapted and applied to analyze datasets in other types of dementia.