集成多种数据源识别导致常见疾病的遗传变异

项目名称： 集成多种数据源识别导致常见疾病的遗传变异

项目编号： No.60805010

项目类型： 青年科学基金项目

立项/批准年度： 2009

项目学科： 武器工业

项目作者： 江瑞

作者单位： 清华大学

项目金额： 22万元

中文摘要： 本项目研究集成基因组、蛋白质组和相互作用组等多种组学数据识别导致人类复杂疾病的遗传变异的模式识别与机器学习方法。在基金资助下，本项目顺利开展，完成了预期目标，所取得的主要成果包括：(1) 发展了一系列推断基因与人类疾病关联关系的计算方法；(2) 发展了一系列推断蛋白质结构域与人类疾病关联关系的计算方法；(3) 发展了集成蛋白质序列保守性、氨基酸物理化学特性和蛋白质结构域相互作用网络等数据建立氨基酸置换与人类疾病关联关系的计算方法；(4) 发展了识别多个遗传变异上位显性相互作用关系的计算方法。基于这些研究成果，项目组发表了29篇学术论文，其中期刊论文20篇（SCI索引15篇），国际会议论文9篇（EI索引5篇）。项目组还基于这些研究成果开发了一批在线软件，并在基于新一代深度测序研究复杂疾病等方面进行了有益的探索。项目负责人多次访问国际知名大学并多次在国际会议上作口头报告，在国内外同行中形成了一定的学术影响。

中文关键词： 单核苷酸多态；氨基酸置换；复杂疾病；模式识别；机器学习

英文摘要： This project studies pattern recognition and machine learning methods for integrating multiple genomics, proteomics and interactomics data sources for the identification of genetic variants that are associated with human complex diseases. With the support of the grant，the project has successfully (1) developed a series of computational methods for the inference associations between genes and human diseases, (2) developed a series of computational methods for the infererence of relationships between protein structural domains and human diseases, (3) developed computational methods for the inference of associations between amino acid substitutions and human diseases via the integrated use of conservation information of protein sequences and domain-domain interaction networks, and (4) developed statistical methods and algorithms for the identification of epistatic interactions between multiple genetic variants. Based on these results, the project has published 29 papers, including 20 journal publications (15 SCI-indexed) and 9 conferences papers (5 EI-indexed). Besides, the projects has also published a series of online software tools and explored the topic of disease studies using the next generation sequencing data. The principle investigator of the project has also been invited to give talks in several famous universities around the world as well as a few top conferences in the field of bioinformatics and systems biology.

英文关键词： Nonsynonymous single nucleotide polymorphism; Amino acid substitution; Complex disease; Pattern recognition; Machine learning