Novel statistical methods for integrating transcriptomic and proteomic data in GWAS

The objective of the proposed project is to derive novel statistical methods to integrate multi-omics data and pathology data in genome-wide association studies (GWAS) for studying complex phenotypes, with the goal of prioritizing genetic variants and identifying causal genes. First, we will develop novel statistical methods to integrate summary-level omics data and pathology data of diverse populations with GWAS data to prioritize risk genes. Second, we will apply our tools to publicly available xQTL data and the ADSP GWAS data. Third, we will also use the ADSP GWAS summary data to conduct causal analysis of other aging-related phenotypes and AD dementia.

We will first develop novel statistical methods to integrate summary-level xQTL data of multiple populations with GWAS data to test gene associations with complex human diseases. We are interested in studying all complex phenotypes that were profiled for the ADSP samples, especially Alzheimer’s disease (AD) and AD-related complex phenotypes. Especially, our lab has access to the ROS/MAP multi-omics data shared by the Rush Alzheimer’s disease center (http://www.radc.rush.edu/), and GTEx data. All samples in the ROS/MAP study are well-characterized with extensive complex phenotypes profiled, including clinical diagnosis of AD, AD-related complex phenotypes, and psychological phenotypes. GTEx provides transcriptomic data of multiple human tissues. We will leverage multiple omics data profiled from the ROS/MAP study and transcriptomics data profiled from GTEx to learn SNP-omics relations, and then integrate such learned relationships with ADSP data to identify risk genes of complex diseases. We will also validate our findings by using omics and pathology data in the requested data sets.

The purpose of using ADSP data is to increase sample size for testing our derived methods for functional genetic association studies of complex phenotypes, studying the genetic etiology of AD and AD-related phenotypes, and validating our finding by using the omics data from Rush Alzheimer's Disease Center. We are not limited to studying AD only. We are flexible to study any complex phenotypes that are profiled for ADSP samples.

This proposed project is to develop novel statistical methods to integrate summary-level multi-omics data such as transcriptomic, proteomics, and epigenetics, and pathology data, in genome-wide association studies (GWAS) of complex phenotypes, with the goal of identifying causal genes. i) We will develop novel statistical method for integrating summary-level omics data and pathology data with GWAS data. ii) We will apply our tools to publicly available summary-level omics data, omics data from the ROS/MAP study, and ADSP GWAS data for studying AD and AD-related phenotypes. iii) We will conduct causal inference to test the causal relationship between AD and other aging-related phenotypes. We propose to test our proposed methods on the applied genomic analysis data to study complex phenotypes that are profiled for ADSP, including AD, AD-related pathology traits, and related psychological disorders.