Analytics Core

摘要 分析核心将处理和协调多组学和多模态内在表型数据，并提供 先进的计算系统生物学方法和数据集成和分析框架 整个项目。我们的多学科团队在多组学数据分析、数据分析和数据分析方面拥有丰富的专业知识。 协调，计算方法开发，多组学数据整合，遗传关联分析， 神经成像分析、数据可视化、生物统计学和网络分析。核心将支持协作 通过提供经过处理的多组学和多模态表型数据， 分析方法和管道的互操作性。此外，核心小组将与行政协调会密切合作， 管理和共享新生成的数据，并为公众提供数据处理文档 access. U19的总体目标是识别和验证中心相关的纵向外周生物标志物 通过整合纵向多组学和多模式， 来自八个独立队列的表型。为了实现这一目标，协调高质量的处理数据， 先进的计算和系统生物学方法和数据分析管道对整个 项目首先，分析核心将提供高质量和协调的高通量多组学数据， 血液和脑组织以及多模式神经成像、认知和死后神经病理学数据。该数据 将允许在临床、基于社区和 多种族群体和样本。核心还将实施和提供先进的计算系统， 用于高维数据分析的生物学框架，使所有三个项目的研究人员能够执行 数据集成和分析。为此，分析核心将由五个团队组成：组学，成像， 临床，神经病理学，计算和系统生物学团队。分析核心将专注于新的 纵向多组学数据的数据处理和综合分析框架开发 纵向多模态内表型。在目标1中，我们将组装并提供最先进的管道， 处理来自血液和脑组织的高维多组学和多模式神经成像数据，以及 作为所有项目的临床、认知和神经病理学表型。在目标2中，我们将实施并提供 先进的多组学和多模态内在表型数据集成和分析方法和管道 为整个项目，其中包括但不限于（1）新的系统生物学和深度学习平台， 识别与大脑中的分子标记相关的血液分子标记;（2） 用于多组学数据整合以预测疾病进展的网络引导的综合分析方法;以及 (3)一个先进的系统查询，分析和可视化系统生物学。核心小组将与 其他核心和项目，以实现对每个组件的多组学和AD表型数据的分析 具体需求。

Abstract The Analytic Core will process and harmonize multi-omics and multimodal endophenotype data, and provide advanced computational systems biology methods and frameworks for data integration and analysis for the entire project. Our multidisciplinary team has substantial expertise in multi-omics data analysis, data harmonization, computational method development, multi-omics data integration, genetic association analysis, neuroimaging analysis, data visualization, biostatistics, and network analysis. The Core will support collaborative work of the U19 project by providing processed multi-omics and multimodal phenotype data and enhancing the interoperability of analysis methods and pipelines. Furthermore, the Core will work closely with the Administrative Core to curate and share newly generated data and to provide documentation of data processing for public access. The overall goal of the U19 is to identify and validate centrally-linked longitudinal peripheral biomarkers of Alzheimer’s disease (AD) in multi-ethnic populations by integrating longitudinal multi-omics and multimodal phenotypes from eight independent cohorts. To achieve this goal, harmonized high-quality processed data and advanced computational and systems biology methods and pipelines for data analysis are critical to the entire project. First, the Analytic Core will provide high-quality and harmonized high-throughput multi-omics data from blood and brain tissue and multimodal neuroimaging, cognitive and postmortem neuropathology data. This data will permit analyses connecting peripheral and central changes within and across clinical, community-based and multi-ethnic cohorts and samples. The Core will also implement and provide advanced computational, systems biology frameworks for high-dimensional data analysis, enabling investigators from all three projects to perform data integration and analysis. To this end, the Analytic Core will be comprised of five teams: Omics, Imaging, Clinical, Neuropathology, and Computational and Systems Biology teams. The Analytic Core will focus on new data processing and integrative analysis framework development of longitudinal multi-omics data in relation to longitudinal multimodal endophenotypes. In Aim 1, we will assemble and provide state-of-the-art pipelines for processing high-dimensional multi-omics from blood and brain tissue and multimodal neuroimaging data, as well as clinical, cognition and neuropathology phenotypes for all projects. In Aim 2, we will implement and provide advanced multi-omics and multimodal endophenotype data integration and analysis approaches and pipelines for the entire project, which include but not limited to (1) novel systems biology and deep learning platforms for identification of blood molecular signatures that are associated with molecular signatures in the brain; (2) network-guided integrative analysis methods for multi-omics data integration to predict disease progression; and (3) an advanced system for query, analysis, and visualization in systems biology. The Core will work closely with the other Cores and Projects to enable analysis of multi-omics and AD phenome data for each components’ specific needs.