Integrating multidimensional genomic data to discover clinically-relevant predictive models-Alzheimer's Supplement

点击翻译按钮获取中文摘要

Genomic instability (GIN) is a primary hallmark of aging, which is the greatest known risk factor for both Alzheimer’s disease (AD) and cancer. Because there is no consensus on which measures of GIN are most biologically and clinically relevant, in our parent project we are testing GIN metrics and developing tools for assessing GINs reproducibly across cancer. Our approaches are designed to be technology-, platform-, and disease-agnostic and therefore should also apply to AD. Our focus, thus far, has been on chromosomal instability (CIN, altered chromosome number and structure; e.g., total number of breakpoints, percent of bases with copy number variation, total functional aneuploidy, etc.) and DNA methylation instabilities (DNAm, e.g., CpG island methylator phenotype; CIMP, widespread altered promoter methylation, density of methylated to non-methylated CpGs, etc.). In cancer we and others have shown GIN is linked to disease etiology and progression, response to therapeutics, and is a potential disease biomarker. While AD animal models confirm DNA integrity impacts neuronal development, function, and maintenance and human aging studies further implicate a role for GIN in brain deterioration, GIN’s role in AD is not clear. There is a critical need to evaluate AD-specific GIN, particularly as potential precision therapy targets and early biomarkers defining therapeutic windows. Our interdisciplinary research team has AD, aging, genomic instability, cancer, genomics, and data science expertise and is well positioned to undertake these studies. Our long term research goal is to understand the role of GIN in the context of aging for multiple conditions and how GIN further contributes to disease etiology, progression, and treatment. Here, we propose the first steps towards demonstrating utility of our methodology in additional diseases by applying them to publicly available AD human and mouse data and comparing the resulting GIN profiles to cancer data analyses in our parent award. We hypothesize this will determine the extent and type of CIN (Aim 1) and DNAm instability (Aim 2) in AD. Critically, we will demonstrate how generalizable our methods and gained knowledge are, add AD examples and vignettes to the tools we are developing, and compare GINs across diseases (AD and cancers), species (human and mouse), and with respect to sex and age. Additionally, we will generate genotype and DNAm data from 3xTG-AD mouse hippocampus (AD-relevant brain tissue), tibialis anterior muscle (as a sentinel organ), and plasma (as a circulating factor) to investigate GIN as an AD biomarker. Critically, with this supplement we will demonstrate generalizability of the parent award methods and knowledge by expanding our existing non-AD NHGRI award to have an AD focus. This work will also stimulate additional activity and collaborations in AD and related dementias by providing preliminary data for several future grant proposals targeting the role of GIN in aging as a general disease mechanism, in AD pathogenesis, for drug repurposing and shared etiology studies between cancer and AD, and with respect to the utility of GINs as peripheral or circulating biomarkers.