DiseaseNet Finder: A Systems Medicine Toolkit for Clinical and Translational Rese

描述(申请人提供)：多种复杂的人类疾病(如癌症、糖尿病、精神分裂症等)具有启动和维持疾病进展的相应复杂的多基因基因型。尽管在过去的几十年里，我们在识别与表型有关的关键基因方面取得了很大进展，但我们对复杂疾病的分子表型功能基础的了解还不够深入。由几个蛋白质以一系列方式相互作用的信号通路过于简化，并为多个相互作用的基因产物介导的行为提供了不充分的模型。对越来越有注解的蛋白质-蛋白质相互作用(PPI)网络的严格研究部分揭示了这一点，已变得明显的是，这些典型信号通路中的许多蛋白质与一组不同的附加蛋白质进行“串扰”，并受其调节，其中这种串扰经常以组织和/或疾病特有的方式介导。我们建议开发并向学术和商业研究社区提供一套集成的软件工具，以满足定量PPI网络分析的未满足需求，从而推动实际的翻译研究和验证。工具DiseaseNet Finder将搜索全球PPI网络中的候选疾病子网络并进行评分。它将允许多种高维基因组类型(GWAS、SNP、CNV、蛋白质组、miRNA等)的集成。具有PPI网络，并包括分类工具。该软件的创新方面包括：组合评分、多种数据类型集成、节点和边缘预测工具，以及通过图形用户界面无缝实现的终点分类和定量评分。 公共卫生相关性：复杂的疾病包括许多与环境相互作用的基因的作用。需要增强的计算研究工具来发现生物标记物和了解复杂的疾病机制，以整合正在积累的各种类型的基因组和蛋白质组数据。这将使个性化医疗得到更快的发展。

DESCRIPTION (provided by applicant): Many complex human diseases (e.g. cancer, diabetes, schizophrenia etc.) have correspondingly complex, polygenic genotypes that initiate and sustain disease progression. Despite significant progress over the past few decades identifying genes critical to mediating phenotype, our understanding of the functional basis of molecular phenotype for complex diseases is insufficient. Signaling pathways that consist of a few proteins interacting in a serial fashion oversimplify and provide inadequate models for the behavior mediated by multiple interacting gene products. Partly revealed by rigorous studies of increasingly well-annotated protein-protein interaction (PPI) networks, it has become clear that many of the proteins in these canonical signaling pathways engage in "crosstalk" with, and are modulated by, an ontologically diverse set of additional proteins, where this crosstalk is frequently mediated in a tissue and/or disease specific manner. We propose to develop and deliver an integrated suite of software tools to the academic and commercial research community to fulfill the unmet demand for quantitative PPI network analysis that can drive practical translational research and validation. The tool DiseaseNet Finder will search for and score candidate disease sub- networks within global PPI networks. It will permit integration of multiple high- dimensional -omic types (GWAS, SNP, CNV, proteomic, miRNA etc.) with PPI networks and include classification tools. Novel aspects of the software include: combinatorial scoring, multi data type integration, node and edge prediction tools, with end-point classification and quantitative scoring seamlessly implemented through graphical user interfaces. PUBLIC HEALTH RELEVANCE: Complex diseases include the contributions of many genes interacting with the environment. Enhanced computational research tools to discover biomarkers and understand complex disease mechanisms are needed to integrate the various types of genomic and proteomics data that are accumulating. This will permit a more rapid development of personalized medicine.