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: | Hong-Hee Won |
Institution: | Sungkyunkwan University |
Department: | SAIHST |
Country: | |
Proposed Analysis: | We aim to develop prediction models for Alzheimer's disease (AD)-related traits such as cognitive decline and clinical progression using deep neural networks. We hypothesized that gene expression levels in brain tissues are informative for such predictions. Given the unavailability of measured gene expression data in brain tissues for the ADNI samples, gene expression levels in AD-related cells/tissues will be predicted using the PrediXcan software. In addition to predicted gene expression levels, we will utilize various data sources as predictor variables, including imaging features, polygenic risk scores, and other covariates (APOE, age, sex, and principal components). We will evaluate the contribution of predicted gene expression levels to the prediction models by comparing the performance of the models with or without gene expression levels. For example, we will evaluate whether predicted gene expression levels in the hippocampus increase the prediction performance for cognitive decline. Our analysis may suggest gene expression levels as novel independent predictors and provide an improved prediction model. |
Additional Investigators | |
Investigator's Name: | Sang Hyuk Jung |
Proposed Analysis: | We aim to develop prediction models for Alzheimer's disease (AD)-related traits such as cognitive decline and clinical progression using deep neural networks. We hypothesized that gene expression levels in brain tissues are informative for such predictions. Given the unavailability of measured gene expression data in brain tissues for the ADNI samples, gene expression levels in AD-related cells/tissues will be predicted using the PrediXcan software. In addition to predicted gene expression levels, we will utilize various data sources as predictor variables, including imaging features, polygenic risk scores, and other covariates (APOE, age, sex, and principal components). We will evaluate the contribution of predicted gene expression levels to the prediction models by comparing the performance of the models with or without gene expression levels. For example, we will evaluate whether predicted gene expression levels in the hippocampus increase the prediction performance for cognitive decline. Our analysis may suggest gene expression levels as novel independent predictors and provide an improved prediction model. |
Investigator's Name: | Soyeon Kim |
Proposed Analysis: | We aim to develop prediction models for Alzheimer's disease (AD)-related traits such as cognitive decline and clinical progression using deep neural networks. We hypothesized that gene expression levels in brain tissues are informative for such predictions. Given the unavailability of measured gene expression data in brain tissues for the ADNI samples, gene expression levels in AD-related cells/tissues will be predicted using the PrediXcan software. In addition to predicted gene expression levels, we will utilize various data sources as predictor variables, including imaging features, polygenic risk scores, and other covariates (APOE, age, sex, and principal components). We will evaluate the contribution of predicted gene expression levels to the prediction models by comparing the performance of the models with or without gene expression levels. For example, we will evaluate whether predicted gene expression levels in the hippocampus increase the prediction performance for cognitive decline. Our analysis may suggest gene expression levels as novel independent predictors and provide an improved prediction model. |
Investigator's Name: | Beomsu Kim |
Proposed Analysis: | We aim to develop prediction models for Alzheimer's disease (AD)-related traits such as cognitive decline and clinical progression using deep neural networks. We hypothesized that gene expression levels in brain tissues are informative for such predictions. Given the unavailability of measured gene expression data in brain tissues for the ADNI samples, gene expression levels in AD-related cells/tissues will be predicted using the PrediXcan software. In addition to predicted gene expression levels, we will utilize various data sources as predictor variables, including imaging features, polygenic risk scores, and other covariates (APOE, age, sex, and principal components). We will evaluate the contribution of predicted gene expression levels to the prediction models by comparing the performance of the models with or without gene expression levels. For example, we will evaluate whether predicted gene expression levels in the hippocampus increase the prediction performance for cognitive decline. Our analysis may suggest gene expression levels as novel independent predictors and provide an improved prediction model. |