Global Infrastructure for Collaborative High-throughput Cancer Genomics Analysis

协作高通量癌症基因组分析的全球基础设施

• 批准号: 9355157
• 负责人: GAD A GETZ
• 金额: $ 94.19万
• 依托单位: BROAD INSTITUTE, INC.
• 依托单位国家: 美国
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-09-20 至 2021-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9355157
• 关键词: Address; Algorithms; Attention; Bioinformatics; Biological; Biomedical Research; Cancer Center; Cancer Gene Mutation; Categories; Classification; Clinic; Clinical; Clinical Data; Clinical Trials; Code; Collaborations; Communities; Complex; Consensus; Correlative Study; Credentialing; Custom; Data; Data Aggregation; Data Analyses; Data Analytics; Data Set; Databases; Deposition; Development; Documentation; Ensure; Evolution; Freezing; Future; General Population; Genome Data Analysis Center; Genome Data Analysis Network; Genomic Data Commons; Goals; Institutes; Knowledge; Link; Malignant Neoplasms; Manuscripts; Mission; Modeling; Molecular; Morphologic artifacts; National Cancer Institute; Paper; Pathway Analysis; Patient-Focused Outcomes; Patients; Phase; Process; Production; Publications; Publishing; Reporting; Reproducibility; Research Infrastructure; Research Personnel; Running; Sampling; Science; Scientist; Services; Structure; Suggestion; Summary Reports; System; The Cancer Genome Atlas; Time; TimeLine; Tumor Subtype; Update; Vertebral column; Work; Writing; base; cancer genome; cancer genomics; clinical diagnostics; clinically relevant; cost; disorder subtype; experience; experimental study; genome analysis; high standard; improved; innovation; insight; member; molecular subtypes; operation; prevent; repository; tool; treatment response; user-friendly; whole genome; working group

Abstract The Cancer Genome Atlas (TCGA) set the standards for large-scale cancer genome projects worldwide. In the next phase, the National Cancer Institute and its Center for Cancer Genomics are planning large-scale projects closely tied to clinical questions and trials. In order to perform the analysis of these data, the NCI is creating a Genome Data Analysis Network (GDAN) of different types of Genome Data Analysis Centers (GDACs). Central to this Network is a single Processing GDAC, which will take all the harmonized data, as stored in the NCI's Genomics Data Commons, and perform higher level integrated analyses on these data to support both the Analysis Working Groups (AWGs) within the Network (which will be formed for each project to perform special analyses of the data and write manuscripts) as well as the entire biomedical research community. Herein we propose to build the centralized Processing GDAC on top of our FireCloud platform, an infrastructure to run large scale computation on the cloud in a fully rigorous and reproducible fashion. FireCloud development was based on our experience with Firehose, the Broad internal platform on which the standard TCGA data and analyses currently run. We propose to create and operate the GDAN Standard Workflow, incorporating tools actively developed and used within the GDAN and across the entire field, with particular emphasis on clinical tools. This Workflow will serve as the starting point for AWGs and set the highest standards of transparency, reproducibility and rigor for cancer genome analysis. The results of the Standard Workflow will be stored in a public database, and accessible via standard APIs, and used together with a continuously updated database of prior knowledge to create scientific reports that will be made available to the community, in a pre-publication manner. Finally, a major innovation is that AWG members will be able to login into FireCloud and rerun the entire workflow, or parts of it, with their own parameters and subsets of the data – thus making the entire GDAN analysis fully reproducible and scalable. Our goals are therefore: (1) To create a global infrastructure for collaborative extreme- scale cancer analysis; (2) Operate the Standard Workflows at scale; (3) Rapidly and continuously evolve the Standard Workflows; and (4) created improved capabilities for reporting, exploring the results, clinical diagnostics and reproducibility.

摘要 癌症基因组图谱(TCGA)为大规模癌症基因组制定了标准 世界各地的项目。在下一阶段，国家癌症研究所及其研究中心 癌症基因组学正在计划与临床问题密切相关的大型项目和 审判。为了对这些数据进行分析，NCI正在创建基因组数据 不同类型的基因组数据分析中心的分析网络(GDAN)。 这个网络的核心是一个单一的处理GDAC，它将承担所有协调的 数据，存储在NCI的基因组数据共享中，并执行更高级别的集成 对这些数据进行分析，以支持以下两个分析工作组 网络(将为每个项目组成，以执行数据的特殊分析和 撰写手稿)以及整个生物医学研究社区。 在此，我们建议在我们的FireCloud之上构建集中处理GDAC 平台，一个在云上运行大规模计算的基础设施，完全严格 和可复制的时尚。FireCloud开发基于我们在 FireHose，标准TCGA数据和分析的广阔内部平台 目前正在运行。我们建议创建和运行GDAN标准工作流程， 纳入在GDAN内部和整个 领域，特别强调临床工具。此工作流将作为开始 为特设工作组设定最高的透明度、再现性和严格性标准 癌症基因组分析。标准工作流的结果将存储在公共 数据库，并可通过标准API访问，并与连续 更新先验知识数据库，以创建将提供的科学报告 以出版前的方式向社区发布。最后，一项重大创新是AWG 成员将能够登录到FireCloud并重新运行整个或部分工作流， 使用它们自己的参数和数据子集-从而使整个GDAN分析 完全可复制和可扩展。 因此，我们的目标是：(1)为协作Extreme创建全球基础设施- 规模癌症分析；(2)在规模上操作标准工作流程；(3)快速和 不断发展标准工作流；以及(4)创建了改进的功能 报告、探索结果、临床诊断和重复性。