There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
| Principal Investigator | |
| Principal Investigator's Name: | Hongyu Zhao |
| Institution: | Yale University |
| Department: | Biostatistics |
| Country: | |
| Proposed Analysis: | A major goal of human genetics studies of complex disease is to develop individual-level genetic risk prediction models and accurate prediction models will have great impacts on disease prevention and treatment strategies. However, despite the identifications of thousands of genetic variants associated with human diseases through genome wide association studies (GWAS), prediction accuracy remains moderate for most diseases including Alzheimer’s disease, which is largely due to the challenges in identifying all the disease-associated variants and accurately estimating their effect sizes, as well as the modest effect sizes of almost all variants identified by GWAS to date. While no analytic strategy can increase those effect sizes, it may be possible to address the challenges of incorporating all disease-associated variants and accurately estimating their effect sizes. Recent advancements in integrative genomic functional annotation, coupled with the rich collection of summary statistics from GWAS, have enabled increase of statistical power in several different settings. We recently proposed a principled framework, AnnoPred [Hu et al. 2017], that incorporates diverse types of annotation data to improve genetic risk prediction. Our framework requires GWAS summary statistics (not individual-level genotype data) as the input training data, and adaptively estimates variant effect sizes through explicit modeling of linkage disequilibrium and functional annotations. Specifically, we first estimate per-SNP heritability using GWAS summary statistics, LD matrix from a reference panel, and pre-defined functional annotations. Each per-SNP heritability estimate quantifies the SNP’s contribution to disease risk and is later used as an informative prior in a Bayesian framework to better estimate the true effect size of each SNP. Genetic risk scores are then calculated as the linear product of the testing sample’s genotypes and effect size estimates. We have achieved consistently though modestly improved prediction accuracy for many complex traits, including breast cancer, Crohn’s disease, rheumatoid arthritis, type-II diabetes, celiac disease, and coronary artery disease. For example, the AUCs of our proposed method are 0.67, 0.70, and 0.63 for breast cancer, Crohn’s disease, and type-II diabetes, while the best AUCs of existing methods based on summary data are 0.63, 0.69, and 0.61. We are requesting the access of the individual level genotype and phenotype data from ADNI and other genome wide association studies of Alzheimer’s disease to both evaluate the performance of the AnnoPred approach in comparison to other genetic risk prediction models, and develop a comprehensive risk model for Alzheimer’s disease. We will further consider the performance for males and females separately to see whether there is a need for sex-specific risk prediction models. Reference: Hu Y, Lu Q, Powles R, Yao X, Yang C, Fang F, Xu X, Zhao H (2017) Leveraging functional annotations in genetic risk prediction for human complex diseases. PLoS Comput Biol 13: e1005589. |
| Additional Investigators | |
| Investigator's Name: | Paul Crane |
| Proposed Analysis: | Dr. Crane will work with Dr. Zhao on the analysis. |
| Investigator's Name: | Qiongshi Lu |
| Proposed Analysis: | Dr. Lu will work with Dr. Zhao on the analysis. |
| Investigator's Name: | Shubhabrata Mukherjee |
| Proposed Analysis: | Dr. Mukherjee will work with Dr. Zhao on the analysis. |
| Investigator's Name: | Wei Jiang |
| Proposed Analysis: | Dr. Mukherjee will work with Dr. Zhao on the analysis. |
| Investigator's Name: | Yixuan Ye |
| Proposed Analysis: | Ms. Ye will work with Dr. Zhao on the analysis. |
