Statistical methods for integromics discoveries

整合组学发现的统计方法

• 批准号: 8131525
• 负责人: XINGHUA LU
• 金额: $ 32.62万
• 依托单位: UNIVERSITY OF PITTSBURGH AT PITTSBURGH
• 依托单位国家: 美国
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-09-01 至 2014-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8131525
• 关键词: Statistical Methods

DESCRIPTION (provided by applicant): Contemporary systems biology is shifting the paradigm of biomedical research from minimalistic studies of individual genes/proteins to integration of information at systems level. Current high throughput biotechnologies enable collection of a large amount of biological information, and the different aspects of the cellular systems are reflected with heterogeneous data, e.g., genomics, epigenomics, transcriptomics and metabolomics. However, it remains a major challenge to systematically integrate this body of information and derive biological insights at a mechanistic level. The overarching goal of this project is to develop a computational system that enables integration of various high throughput "omics" data (an "integromics" approach) to gain insights into cellular systems, in particular the signal transduction systems. The activities of the project are organized into four specific aims, which progress from approaches for capturing general information among the multiple omics data to more specific and complex models designed to decipher specific cellular signaling systems. Firstly, we will develop a general framework, based on information theory and probabilistic models, to identify information modules that convey biological information between different "omics" data at large scale. Secondly, we develop methods to further investigate if the information from the multiple omics data reflects causal relationships. Thirdly, we will develop tools to recover missing information from the system to augment the high throughput technologies. Finally, we will develop a unified model to elucidate signal transduction pathways by integrating information form multiple omics data in manner that is both biologically sensible and mathematically rigorous. We expect that the methodologies developed in the project are widely applicable to study a variety of cellular signal transduction systems.

描述（由申请人提供）： 当代系统生物学正在将生物医学研究的范式从单个基因/蛋白质的简约研究转变为系统水平的信息整合。当前的高通量生物技术能够收集大量生物信息，并且细胞系统的不同方面通过异构数据反映，例如基因组学、表观基因组学、转录组学和代谢组学。然而，系统地整合这些信息并在机械层面上获得生物学见解仍然是一个重大挑战。该项目的总体目标是开发一种计算系统，能够整合各种高通量“组学”数据（“整合组学”方法），以深入了解细胞系统，特别是信号转导系统。该项目的活动分为四个具体目标，从捕获多个组学数据中的一般信息的方法发展到旨在破译特定细胞信号系统的更具体和复杂的模型。首先，我们将开发一个基于信息论和概率模型的通用框架，以识别在大规模不同“组学”数据之间传递生物信息的信息模块。其次，我们开发了方法来进一步调查来自多个组学数据的信息是否反映了因果关系。第三，我们将开发工具来恢复系统中丢失的信息，以增强高吞吐量技术。最后，我们将开发一个统一的模型，通过以生物学上合理且数学上严格的方式整合来自多个组学数据的信息来阐明信号转导途径。我们期望该项目开发的方法能够广泛适用于研究各种细胞信号转导系统。