GenePattern Notebooks for Cancer Research

 描述（由申请人提供）：在过去的十年中，癌症研究的景观已经改变了公开可用和研究人员生成的数据集的爆炸，以及快速增长的复杂计算方法和工具来整合和分析它们。研究界面临着持续的挑战，因为它试图利用这些丰富的数据和分析工具来推动癌症研究议程。整个癌症研究社区需要一种方式来轻松地协作，记录，捕获和共享他们的工作，从概念到分析到出版。此外，癌症生物学家可能难以选择正确的工具并正确使用它们，从而有效地将这种强大的功能置于他们的直接接触之外。U24提案的目标是使用GenePattern计算基因组学平台，该平台自2004年以来一直服务于癌症社区，作为满足这些需求的新电子笔记本环境的基础。通过这些努力，我们将支持处于癌症研究前沿的多样化用户社区，他们寻求更好地了解疾病的潜在机制，将改进的患者诊断和预后方法转化为临床，并确定新的药物靶点。 目标1.开发一个GenePattern电子笔记本，用于计算机协作研究。利用GenePattern、Google Drive/Google Cloud和IPython平台的新组合，我们将开发一个用于创建和部署电子笔记本的环境，以支持整个正在进行的协作研究，包括运行分析、呈现结果、记录评论和解释结果，以及捕获可重现的计算工作流程。 目标2.创建一个用于癌症研究的GenePattern笔记本集合。我们将制定和部署动态GenePattern笔记本，体现基于驱动癌症项目的完整分析研究，以指导研究人员在每个分析执行步骤中进行相关考虑，以最好地支持他们的研究目标。 目标3.添加GenePattern模块以解决癌症的复杂性。我们将根据目标2中笔记本收集的需要添加新模块，包括识别生物标志物、聚类、分类和降维的新信息理论方法。 目标4。为癌症研究社区提供培训和GenePattern Notebook支持。我们将为笔记本电脑环境提供高水平的支持;开发以癌症为重点的培训材料，以基于推动癌症项目的笔记本电脑为特色;在匹兹堡超级计算中心的高性能计算基础设施上部署公共基因模式服务器。

 DESCRIPTION (provided by applicant): Over the past decade, the landscape of cancer research has changed with the explosion of publicly available and investigator generated datasets, and the rapidly growing number of sophisticated computational methods and tools to integrate and analyze them. There are continuing challenges to the research community as it seeks to harness this wealth of data and analysis tools to move the cancer research agenda forward. The entire cancer research community needs a way to easily collaborate on, document, capture, and share their work, from conception through analysis to publication. Moreover, cancer biologists may have difficulty choosing the right tools and using them correctly, effectively putting this powerful capability out of their direct reach. The goal of thisU24 proposal is to use the GenePattern computational genomics platform, which has served the cancer community since 2004, as the foundation for a new electronic notebook environment to meet these needs. Through these efforts we will support a diverse community of users at the forefront of cancer research who seek to better understand the underlying mechanisms of disease, translate improved methods for patient diagnosis and prognosis to the clinic, and identify new drug targets. Aim 1. Develop a GenePattern electronic notebook for collaborative in silico research. Leveraging a novel blend of GenePattern, Google Drive/Docs, and the IPython platform, we will develop an environment for creating and deploying electronic notebooks to support the entirety of ongoing collaborative studies, including running analyses, presenting results, recording comments and interpretation of results, and capturing the reproducible computational workflow. Aim 2. Create a collection of GenePattern notebooks for cancer research. We will formulate and deploy dynamic GenePattern notebooks embodying complete analysis studies based on driving cancer projects, to guide investigators through relevant considerations at each analysis execution step to choices best supporting their research goals. Aim 3. Add GenePattern modules to address cancer complexity. We will add new modules as required for the notebook collection in Aim 2, including new information-theoretic approaches to identifying biomarkers, clustering, classification, and dimension reduction. Aim 4. Provide training and GenePattern Notebook support for the cancer research community. We will provide a high level of support for the notebook environment; develop cancer focused training materials featuring notebooks based on driving cancer projects; deploy a public GenePattern server on the high- performance computing infrastructure at the Pittsburgh Supercomputing Center.