Skip to navigation Skip to main content Skip to footer

Polygenic effects of Alzheimer's Disease associated genetic risk factors in the early life

Polygenic effects of Alzheimer's Disease associated genetic risk factors in the early life

Principal Investigator: Professor Jun Zhu
Approved Research ID: 33463
Approval date: March 4th 2020

Lay summary

Late-Onset Alzheimer's Disease (LOAD) is a neurodegenerative disease known to manifest in the old age. Its prevalence rate also continues to rise with the increased lifespan in the general population. However, its pathological changes may begin earlier in life and its understanding is critical in the development of prevention plans as well as early treatment. Our research seeks to understand AD in order to identify clinically relevant biomarkers and prognostic measures, and the function of disease associated risk variants at the phenome level.

The strategy taken here is rooted in the current understanding of genetic risks for AD. Leveraging the already known fact that single variant alone is insufficient to explain AD, we aim to study the effect of multiple genetic effects together. Assuming that common AD risk variants may excert small effects toward cognitive decline, we hypothesize that continuous decrease in the cognitive functions will be present with increasing polygenic risk for AD. We also hypothesize that this relationship will be mediated by the changes in the brain connectome and may be traceable by metabolic markers in the blood.

The findings here will be valuable to translational research in two folds. One is that we can test the traits associated with polygenic effect in the longitudinal AD cohort in order to validate its efficacy in early disease detection. Furthermore, its mechanistic findings can be mapped onto specific genes that can be further tested in the post-mortem brain tissues of AD patients that is available here at Mount Sinai. Second is that these findings may provide mechanistic understanding of healthy cognitive aging at the brain connectome level.

The duration of this work may span from 2-3 years. The pipeline and analysis being used in this study utilizes bayesian methods which is hypothesis driven, yet it consumes a lot of computation time and power. Although we have access to >500 nodes of quad-core clusters, it is ultimately a shared resource within our institution and we safely expect to complete the data generation within a year after adequate quality control procedures. The basic analysis may take upto 3 months and network analysis may take upto 6 months to 1 year. The writing of manuscript may take upto 3 months.