Statistical Methods for Network-Based Integrative Analysis of CVD Epigenetic Data

 描述(由申请人提供)：该项目涉及开发新的统计方法和计算工具，用于基于网络的心血管疾病表观遗传风险因素的综合分析。虽然来自新技术的组学数据的出现促进了表观遗传因素的研究，但现有的方法往往没有考虑到生物数据的复杂性，例如由于作为生物途径一部分的基因/蛋白质的相互作用而导致的相关性，并且无法科学地整合不同的组学数据 例如遗传变异、DNA甲基化和基因表达。本项目中提出的方法以及将为实施这些方法而开发的软件工具，解决了这些缺点，并促进了生物医学界的进一步研究，以更好地了解心血管疾病的潜在生物学，并为治疗开发新的诊断生物标记物和潜在靶点。拟议的方法是基于对动脉粥样硬化多人种研究(MESA)的表观遗传学数据的研究，包括(1)网络途径丰富分析方法，它纳入了基因和蛋白质之间相互作用的现有知识，同时补充和重新fi这些信息(目标1A)，及其扩展用于分析多种类型的组学数据(目标1B)；(2)综合分析框架，以确定基因表达水平和DNA甲基化之间的联系(目标2A)，并通过综合分析DNA甲基化和基因表达数据确定多种心血管疾病表观遗传学共同的表观遗传因素(目标2B)。我们将为拟议的方法(目标3)开发ffi熟知和用户友好的软件工具，这些工具将在使用模拟数据和MESA真实数据进行广泛测试后向公众免费提供。

 DESCRIPTION (provided by applicant): This project involves the development of new statistical methodologies and computational tools for network-based integrative analysis of epigenetic risk factors of cardiovascular diseases (CVD). While the advent of omics data from new technologies has facilitated the study of epigenetic factors, existing methodologies often do not account for complexities of biological data such as correlations due to interactions of genes/proteins as part of biological pathways and fail to efficiently integrate diverse omics data sets for instance genetic variation, DNA methylation and gene expression. The methodologies proposed in this project, and the software tools that will be developed to implement them, address these shortcomings, and facilitate further research by the biomedical community to gain a better understanding of the underlying biology of CVD, and to develop new diagnostic biomarkers and potential targets for therapies. The proposed methodologies are motivated by the study of epigenetic data from the Multi-Ethnic Study of Atherosclerosis (MESA), and include (i) a network-based pathway enrichment analysis method that incorporates available knowledge of interactions among genes and proteins while complementing and refining such information (Aim 1A), as well as its extension for analysis of multiple types of omics data (Aim 1B), and (ii) an integrative analysis framework to identify associations among gene expression levels and DNA methylation (Aim 2A) and identify common epigenetic factors of multiple CVD phenotypes through integrated analysis of DNA methylation and mRNA expression data (Aim 2B). We will develop efficient and user-friendly software tools for the proposed methods (Aim 3), which will be made freely available to the public after extensive tests using both simulated data, as well as real data from MESA.