U01AG073079

Causal and Integrative Deep Learning for Alzheimer's Disease Genetics - Summary

In response to PAR-19-269, “Cognitive Systems Analysis of Alzheimer's Disease Genetic and Phenotypic Data”, we propose developing and applying more powerful and robust machine learning methods for causal and integrative analysis, especially deep learning approaches for instrumental variable analysis, to identify causal risk/protective factors for Alzheimer's disease (AD) in the post-GWAS era.

By leveraging published large-scale GWAS, whole-genome sequencing (WGS), and other omic and neuroimaging data, our main motivation is to extend an emerging and increasingly influential approach of integrating GWAS with gene expression data, called transcriptome-wide association studies (TWAS).

TWAS aims to improve over the current practice of GWAS by not only increasing statistical power but also identifying (putative) causal genes, thus gaining insights into the genetic basis of common diseases and complex traits. The statistical principle underlying TWAS is the (two-sample) two-stage least squares (2SLS) for linear models in the framework of instrumental variable (IV) analysis for causal inference.

In practice, however, TWAS may fail to identify true causal genes while giving false positives due to the violation of its modeling assumptions, e.g., due to non-linear effects of IVs or gene expression, or due to invalid IVs (in the presence of horizontal pleiotropy of SNPs).

First, we propose developing linear models and neural network models incorporating a large number of functional annotations on the genome (e.g., various types of functional genomic and epigenetic data from the ENCODE and Roadmap Epigenomics projects) as prior knowledge to improve imputing/predicting gene expression (or other molecular or imaging endophenotypes or complex traits/diseases) via SNPs, corresponding to the first stage of 2SLS.

Second, we propose neural networks as more flexible non-linear models for the second stage of 2SLS in the presence of invalid IVs, which may be the SNPs having direct (or horizontal pleiotropic) effects on the outcome as expected from the widespread pleiotropy. Then we combine the approaches in the above two stages to form a more flexible and robust neural network approach as an extension of 2SLS for causal inference.

Third, we consider inferring causal directions between two traits, e.g., a gene's expression and AD, allowing non-linear relationships between SNPs and traits and between the two traits. This is critical in reducing false positives, e.g., due to reverse causation, but has been largely under-studied.

Fourth, we apply the new (and existing) methods to transcriptomic, proteomic, neuroimaging, and AD GWAS/WGS data to identify (putative) causal genes, proteins, and brain regions of interest (ROIs) for AD, while building the corresponding genetic prediction models for endophenotypes and AD risk.

Finally, we will develop and disseminate publicly available software implementing the proposed analysis methods, e.g., as Python programs or R packages, to facilitate the wide use by the scientific community.

Regents Of The University Of Minnesota

TO ENCOURAGE BIOMEDICAL, SOCIAL, AND BEHAVIORAL RESEARCH AND RESEARCH TRAINING DIRECTED TOWARD GREATER UNDERSTANDING OF THE AGING PROCESS AND THE DISEASES, SPECIAL PROBLEMS, AND NEEDS OF PEOPLE AS THEY AGE. THE NATIONAL INSTITUTE ON AGING HAS ESTABLISHED PROGRAMS TO PURSUE THESE GOALS. THE DIVISION OF AGING BIOLOGY EMPHASIZES UNDERSTANDING THE BASIC BIOLOGICAL PROCESSES OF AGING. THE DIVISION OF GERIATRICS AND CLINICAL GERONTOLOGY SUPPORTS RESEARCH TO IMPROVE THE ABILITIES OF HEALTH CARE PRACTITIONERS TO RESPOND TO THE DISEASES AND OTHER CLINICAL PROBLEMS OF OLDER PEOPLE. THE DIVISION OF BEHAVIORAL AND SOCIAL RESEARCH SUPPORTS RESEARCH THAT WILL LEAD TO GREATER UNDERSTANDING OF THE SOCIAL, CULTURAL, ECONOMIC AND PSYCHOLOGICAL FACTORS THAT AFFECT BOTH THE PROCESS OF GROWING OLD AND THE PLACE OF OLDER PEOPLE IN SOCIETY. THE DIVISION OF NEUROSCIENCE FOSTERS RESEARCH CONCERNED WITH THE AGE-RELATED CHANGES IN THE NERVOUS SYSTEM AS WELL AS THE RELATED SENSORY, PERCEPTUAL, AND COGNITIVE PROCESSES ASSOCIATED WITH AGING AND HAS A SPECIAL EMPHASIS ON ALZHEIMER'S DISEASE. SMALL BUSINESS INNOVATION RESEARCH (SBIR) PROGRAM: TO EXPAND AND IMPROVE THE SBIR PROGRAM, TO INCREASE PRIVATE SECTOR COMMERCIALIZATION OF INNOVATIONS DERIVED FROM FEDERAL RESEARCH AND DEVELOPMENT, TO INCREASE SMALL BUSINESS PARTICIPATION IN FEDERAL RESEARCH AND DEVELOPMENT, AND TO FOSTER AND ENCOURAGE PARTICIPATION OF SOCIALLY AND ECONOMICALLY DISADVANTAGED SMALL BUSINESS CONCERNS AND WOMEN-OWNED SMALL BUSINESS CONCERNS IN TECHNOLOGICAL INNOVATION. SMALL BUSINESS TECHNOLOGY TRANSFER (STTR) PROGRAM: TO STIMULATE AND FOSTER SCIENTIFIC AND TECHNOLOGICAL INNOVATION THROUGH COOPERATIVE RESEARCH DEVELOPMENT CARRIED OUT BETWEEN SMALL BUSINESS CONCERNS AND RESEARCH INSTITUTIONS, TO FOSTER TECHNOLOGY TRANSFER BETWEEN SMALL BUSINESS CONCERNS AND RESEARCH INSTITUTIONS, TO INCREASE PRIVATE SECTOR COMMERCIALIZATION OF INNOVATIONS DERIVED FROM FEDERAL RESEARCH AND DEVELOPMENT, AND TO FOSTER AND ENCOURAGE PARTICIPATION OF SOCIALLY AND ECONOMICALLY DISADVANTAGED SMALL BUSINESS CONCERNS AND WOMEN-OWNED SMALL BUSINESS CONCERNS IN TECHNOLOGICAL INNOVATION.

93.866 - Aging Research

National Institute on Aging (NIA) [HHS - NIH]

Minneapolis, Minnesota 554143063 United States

Single Zip Code

Cognitive Systems Analysis of Alzheimer's Disease Genetic and Phenotypic Data (U01 Clinical Trial Not Allowed) (PAR-19-269)

Amendment Since initial award the total obligations have increased 376% from $733,352 to $3,492,892.