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: | Yize Zhao |
| Institution: | Yale University |
| Department: | Biostatistics |
| Country: | |
| Proposed Analysis: | The overarching goal of this project is to develop and evaluate a set of scalable and biologically meaningful statistical learning methods via leveraging the power of large-scale imaging phenotypes to identify new genetic markers for Alzheimer's disease (AD) risk and progression as well as quantify the pathological mechanism among genetics, brain activity and disease behavior. AD is a national priority with no available cure, and there is an urgent need for developing effective strategies to identify new AD genetic risk or protective factors for disease modeling and drug development. Compared with categorical diagnoses, imaging quantitative trait (QT) has distinct advantages to capture disease etiology and improve identification power. However, existing analytical methods for neuroimaging genetics studies suffer from serious limitations due to a) high-dimensionality in both imaging and genetics data, b) unique biological structure information within both imaging and genetics data, and c) complex pathological mechanisms among imaging, genetics and disease phenotypes. To overcome these barriers, this project proposes the following four aims: 1) to establish robust and biologically driven heritability analyses to jointly analyze whole brain neuroimaging biomarkers of AD risk and progression, 2) to conduct multi-trait fine mapping to identify pleiotropic genotypes for AD risk and progression under scalable Bayesian variable selection algorithm, 3) to investigate the fundamental pathological mechanism from the identified genetic variants to AD risk and progression jointly mediated by the selected neuroimaging endophenotypes, and 4) to perform systematic evaluation of the proposed methods including real data analyses and extensive, realistic simulations, and to develop user-friendly analytical pipelines. Our proposed methods are innovative in four important aspects: I) to make the very first attempt to conduct large-scale joint heritability analysis while accounting for sparsity, brain structural network and different sources of correlation, II) to propose a scalable and structural driven multi-trait fine mapping method based on Bayesian variable selection and shrinkage which has not been investigated before, III) to conduct multi-mediator Bayesian mediation analyses to uncover how genetic variants and neuroimaging traits impact AD risk and progression which has not been studied comprehensively before, and IV) to systematically perform and validate the proposed analyses in two AD cohorts with one of them just released very recently, and to develop efficient and user friendly pipelines which are in great need for biomedical and clinical community. Successful completion of the above aims will produce innovative statistical and computing methods and tools for genetic analysis of neuroimaging and clinical data to address critical barriers in brain imaging genetics. Using the Alzheimer's Disease Neuroimaging Initiative and a local AD cohort as test beds, these methods will be shown to have great potential for understanding the genetic and neurobiological mechanism of AD, and be expected to advance neurological and psychiatric research in general and benefit public health outcomes. |
| Additional Investigators | |
| Investigator's Name: | Xiwen Zhao |
| Proposed Analysis: | Help provide analytical support for the heritability analysis using PET data, including data manipulation, data processing, and simple analysis. |
