A Statistical Computing Framework for Genomic Data

描述（由申请人提供）：我们将设计和部署软件基础设施，桥梁在生物技术和信息科学的不同过程和资源。在此过程中，我们将设计和实施R编程语言的重大改进，以支持我们开发的创新软件工具的开发和使用。在生物技术领域，我们将设计能够整合实验数据和实验元数据的软件（例如MIAME）。我们将开发工具来整合生物元数据与实验数据。这些工具将被设计为允许其他开发人员访问所有方法，并将被设计为通过定义良好的类系统简化与其他项目的交互。我们将开发可视化的软件基础设施，结合来自多个来源的数据和访问信息，从WWW编程。在信息技术领域，将开发用于创建、构建和收集已发表文献的注释资源和数据库的工具。将提供用于将实验结果与真实的注释/文献资源链接的交互式工具。我们将探讨Web服务的开发和部署。软件架构将解决注释和科学文献的高度动态性，并将科普质量不同的多个数据源。将支持按错误率、分辨率和能力编制技术索引。将开发和部署能够以图形方式使用软件工具的指南和基础设施。 此外，我们将使用上述工具开发新的基因组数据的计算方法，特别注意可视化，计算推理，多重比较和专门的分析方法的微阵列数据。

DESCRIPTION (provided by applicant): We will design and deploy software infrastructure that bridges diverse processes and resources in biotechnology and information science. In the process we will design and implement significant improvements in the R programming language to support the development and use of the innovative software tools we develop. In the domain of biotechnology we will design software enabling the integration of experimental data and experimental metadata (e.g. MIAME). We will develop tools to integrate biological metadata with experimental data. These tools will be designed to allow other developers to have access to all methodology and will be designed to simplify interactions with other projects through a well-defined class system. We will develop software infrastructure for visualization, combining data from multiple sources and accessing information, programmatically from the WWW. In the domain of information technology, tools for creating, structuring and harvesting annotation resources and databases of published literature will be developed. Interactive tools for linking experimental results to annotation/literature resources in real time will be provided. We will explore the development and deployment of Web services. The software architecture will address the highly dynamic nature of the annotation and scientific literature, and will cope with multiple data sources of varying degrees of quality. Indexing of techniques with respect to error rates, resolutions and capabilities will be supported. Guidance and infrastructure that will enable the use of the software tools in a graphical manner will be developed and deployed. Additionally we will use the tools described above to develop new computational methods for genomic data, with particular attention to visualization, computational inference, multiple comparisons, and specialized analytic methods for microarray data.