Modeling functional genomics of susceptibility to the persistent effects of environmental toxins in an elderly rural Indiana neurodegenerative cohort

PROJECT SUMMARY Alzheimer's disease and related dementias (ADRD) have become a growing public health crisis and are one of the major causes of disability among people over 65. Even though it was discovered over a century ago, its driving mechanism remains poorly understood, which limiting intervention options. Scientists believe that ADRD

is caused by many genes through complex interactions with environmental factors. Over 70 genetic regions have been identified to be associated with ADRD; however, how these genes contribute to disease penetrance and progression remains elusive. Therefore, a comprehensive characterization of gene regulations at single-cell

resolution is warranted to understand the molecular mechanism and discover therapeutic targets for ADRD. Recent advances in single-cell sequencing technologies, especially the simultaneous measurement of gene expression and chromatin accessibility for the same cell, along with the community efforts to share omics data

provide unprecedented opportunities to unravel the dynamic molecular mechanism of ADRD. To understand the contribution of gene-environment (GxE) interactions and leverage single-cell genomics approaches and pathway analysis, we propose to test the hypothesis that persistent neurotoxicity is due to exposures altering self-

perpetuating homeostatic processes that give the resiliency and perpetuity to the adverse toxicological processes by either genetic and/or epigenetic means in the parent R01. More specifically, we propose to identify genetic pathways associated with establishing a persistent neurotoxic state via single-cell genomics and bioinformatic

comparisons to population level data in Aim 2; and Aim 3 focuses on the validation of the genetic pathways of GxE induction of persistent neurotoxicity with in vivo and in vitro models. To augment the parent R01 and integrate single-cell human brain tissue data from healthy individuals available through The BRAIN Initiative, the overarching goal of the supplement is to identify the network alterations of AD

patients by comparing them to those from healthy human brain tissues. This supplement is built upon our proposed new machine learning tools for network construction by integrating snRNA-seq and snATAC-seq data in the parent R01 and our recent successful demonstration of constructing a genome-wide gene regulatory

network using bulk RNA-seq and SNP data based on a two-stage penalized least square method. Specifically, we will (1) Harmonize the single-cell RNA and ATAC data from the “NeMO” and “SEA-AD” studies, and (2) Conduct transcriptome-wide causal inference on gene regulation for healthy and AD/ADRD cohorts. Results

from the single cell analysis of parent R01 and this supplement provides a comprehensive evaluation of genetic and environmental factors and their complex interactions during the initiation and progression of ADRD, which ultimately provide new insights into the altered gene regulation of ADRD that could provide a foundation for

developing new strategies to prevent, delay, or treat ADRD. Our integrative machine learning tools will be shared through open-source programs, which can benefit single-cell-based research in the broader scientific community.