AD Risk Prediction
- PI
- Engelman, Corinne
- DAR ID
- 10270
- Initial Approval Date
- February 26, 2021
- Current DAR Status
- Active
- PI Institution
- University of Wisconsin - Madison
- PI Institution Country
- United States
- PI Institution Type
- Non-Profit
Requested Datasets (6)
NG00067Alzheimer's Disease Sequencing Project Umbrella StudyNG00113Metabolomic and lipidomic signatures in Alzheimer disease brainsNG00068ADC4 - Alzheimer's Disease Center Dataset 4NG00100Novel Alzheimer Disease Risk Loci and Pathways in African American Individuals Using the African Genome Resources Panel: A Meta-analysis. Kunkle et al. (2021)NG00177ADGC Multi-Ancestry GWAS Meta-Analysis – Rajabli et al., 2025NG00179AD GWAS in participants of African ancestry – Ray et al., 2024
Research Use Statement
Currently, no combined measure of rare (minor allele frequency [MAF] less than 0.5%), low frequency (MAF 0.5% to 5%), and common variant (MAF 5% or higher) risk of Alzheimer’s disease (AD) exists. Common variant polygenic risk scores (PRSs) have been a central approach to predicting genetic risk of AD, however, we know that rare and low frequency variants can account for the missing heritability in AD. One objective of this project is to determine the risk for AD based on variants across the full allele frequency spectrum. An additional objective of this project is to utilize GWAS summary statistics from multiple ancestries to calculate the PRS and AD risk.
To accomplish these objectives, we will leverage sequencing data from the ADSP (our study is contributing 1,531 samples to the Follow Up phase) and summary statistics from published GWAS from different ancestries. To generate the common variant PRS, we will use these summary statistics and methods such as PRS-CSx. We will determine the carrier status for rare and low frequency AD risk alleles based on the literature and bioinformatics tools. Prediction of AD case-control status and age-at-onset for quantiles of the common variant PRS and carrier status will be characterized with an empirical receiver operating characteristic (ROC) curve. The statistical software R will be used to perform regression analyses and to evaluate the AUC.
To accomplish these objectives, we will leverage sequencing data from the ADSP (our study is contributing 1,531 samples to the Follow Up phase) and summary statistics from published GWAS from different ancestries. To generate the common variant PRS, we will use these summary statistics and methods such as PRS-CSx. We will determine the carrier status for rare and low frequency AD risk alleles based on the literature and bioinformatics tools. Prediction of AD case-control status and age-at-onset for quantiles of the common variant PRS and carrier status will be characterized with an empirical receiver operating characteristic (ROC) curve. The statistical software R will be used to perform regression analyses and to evaluate the AUC.
Non-Technical Summary
Currently, risk prediction for the later-onset form of Alzheimer’s disease (AD) focuses on genetic variants that are more common in the population, but ignores less common variants. Prediction is also largely based on data from populations of European ancestry. The goals of this project are to incorporate genetic variants across the full allele frequency spectrum (more and less common genetic variants) and to include more ancestrally diverse populations. To accomplish these goals, we will leverage genetic data from the ADSP and summary statistics (results) from published studies in diverse populations. We will determine the prediction of AD case-control status and age-at-onset.
Reported Publications
Publications shown are those reported by investigators through NIAGADS DAR reporting and renewal processes and may not represent a complete list of publications associated with the project.
No publications have been reported for this DAR.