COMBINED COMPUTATIONAL AND EXPERIMENTAL ANALYSES OF GENE REGULATION BY MICRORNAS

微米基因调控的计算和实验相结合的分析

• 批准号: 8244418
• 负责人: Xiaowei Wang
• 金额: $ 30.1万
• 依托单位: WASHINGTON UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2010
• 资助国家: 美国
• 起止时间: 2010-04-01 至 2015-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8244418
• 关键词: Algorithms; Animals; Bioinformatics; Biological; Biological Process; Cancer Control; Categories; Cells; Cellular Assay; Communities; Computer Simulation; Computing Methodologies; Data; Data Set; Databases; Development; Disease; Gene Expression Regulation; Gene Targeting; Genes; Goals; Human; Internet; Knowledge; Lead; Learning; Machine Learning; MicroRNAs; Modeling; Molecular Profiling; Nature; Oncogenes; Oncogenic; Pathway interactions; Performance; Physiological Processes; Process; Publishing; Regulation; Source; Techniques; Testing; Therapeutic; Time; Training; base; computer framework; computerized tools; design; disorder control; improved; novel; overexpression; performance tests; public health relevance; tumor; wiki

DESCRIPTION (provided by applicant): MicroRNAs (miRNAs) are involved in many diverse biological processes and a single miRNA can modulate the expression of hundreds of gene targets. Identification of miRNA targets is a critical first step for miRNA functional studies. Unfortunately, none of the existing computational tools are sufficient for the accurate prediction of miRNA targets. The major goal of this study is to significantly improve our ability to predict miRNA gene targets by developing and validating a new computational target prediction model. Our preliminary data indicate that machine learning techniques are effective in building improved target prediction models using expression profiling data. Based on these studies, we propose the hypothesis that the application of novel computational methods on large expression profiling datasets will lead to a robust bioinformatics model for miRNA target prediction. Under Specific Aim 1, we will build this model by first performing microarray profiling analyses to identify a large number of miRNA-down regulated genes. The expression profiling data generated from our study will then be used as training data to identify general sequence features that are important to miRNA target prediction. These features will be systematically analyzed in various computational frameworks to build a robust target prediction model, which will later be validated with independent experimental data. Under Specific Aim 2, this new model will be used to develop an online miRNA database for target prediction for all known miRNAs in humans and animals. In addition to bioinformatics target prediction data, the new database will also host data integrated from many heterogeneous sources. Thus, our new database will serve as an integrated web portal to retrieve relevant miRNA functional data with a focus on miRNA target gene regulation. Finally, under Specific Aim 3, we will evaluate the usefulness of the new computational model and database by initiating the development of new miRNA-based therapeutic strategies for disease control. As a specific example, we will test predicted oncogene modulation by natural and artificial miRNAs as a potential new strategy for cancer control. The successful completion of this project will significantly advance our ability to understand the gene regulation by miRNAs in a variety of normal physiological processes and disease states. PUBLIC HEALTH RELEVANCE: MicroRNAs are involved in many diverse biological processes and a single microRNA can modulate the expression of hundreds of gene targets. The major goal of this study is to significantly improve our ability to predict microRNA gene targets by developing and testing a new computational microRNA target prediction model. The successful completion of this project will significantly advance our ability to understand the gene regulation by microRNAs in a variety of normal physiological processes and disease states.

描述（由申请人提供）：微小RNA（miRNA）参与许多不同的生物过程，单个miRNA可以调节数百个基因靶点的表达。miRNA靶点的鉴定是miRNA功能研究的关键第一步。不幸的是，现有的计算工具都不足以准确预测miRNA靶点。本研究的主要目标是通过开发和验证新的计算靶标预测模型来显著提高我们预测miRNA基因靶标的能力。我们的初步数据表明，机器学习技术在使用表达谱数据构建改进的目标预测模型方面是有效的。基于这些研究，我们提出了一个假设，即新的计算方法在大型表达谱数据集上的应用将导致一个强大的生物信息学模型用于miRNA靶点预测。在特定目标1下，我们将首先通过微阵列分析来建立这个模型，以识别大量的miRNA下调基因。从我们的研究中产生的表达谱数据将被用作训练数据，以识别对miRNA靶点预测重要的一般序列特征。这些特征将在各种计算框架中进行系统分析，以建立一个强大的目标预测模型，随后将使用独立的实验数据进行验证。在Specific Aim 2中，这个新模型将用于开发一个在线miRNA数据库，用于预测人类和动物中所有已知miRNA的靶标。除了生物信息学目标预测数据外，新数据库还将托管从许多异构来源整合的数据。因此，我们的新数据库将作为一个综合的门户网站，检索相关的miRNA功能数据，重点是miRNA靶基因调控。最后，在具体目标3下，我们将通过启动新的基于miRNA的疾病控制治疗策略的开发来评估新的计算模型和数据库的有用性。作为一个具体的例子，我们将测试预测的癌基因调节天然和人工的miRNA作为一个潜在的新的癌症控制策略。该项目的成功完成将大大提高我们理解miRNAs在各种正常生理过程和疾病状态下的基因调控的能力。 公共卫生关系：microRNA参与许多不同的生物过程，单个microRNA可以调节数百个基因靶点的表达。本研究的主要目标是通过开发和测试一种新的计算microRNA靶点预测模型来显着提高我们预测microRNA基因靶点的能力。该项目的成功完成将大大提高我们理解microRNA在各种正常生理过程和疾病状态下的基因调控的能力。