Computational methods for human genomic data integration: demography, selection,

描述(由申请人提供)： 最近在数据收集方面的投资已经产生了关于基因组功能、人类变异和哺乳动物进化的丰富的信息目录，但需要改进的计算工具来整合和解释这些目录，以便能够获得新的生物学知识和医学进展。在这里，我们概述了一个雄心勃勃的项目，即开发计算方法，将公共可用的数据目录整合在一起，以提供关于人类基因组中序列的进化和功能的深刻新见解。我们的建议特别关注非编码序列，它们仍然是基因组中注释最稀少、理解最少的区域。该提案解决了三个基本且密切相关的问题：(1)人口统计学推断，为统计遗传学提供改进的“零模型”，并揭示基因流、自然选择和其他现象的局部特征；(2)检测散布在转录因子结合位点等非编码序列上的自然选择，以提供有关其功能和形成它们的进化过程的信息；以及(3)基于整合的数据集对全基因组范围内的“功能潜力”进行预测，以确定新的功能元件，并为实验后续确定候选疾病基因座的优先顺序。我们的建议包括创新的统计建模，新的推理算法，开发免费可用的软件工具和浏览器资源，以及对最新基因组数据集的详细分析。据我们所知，这将是迄今为止进行的最全面的努力，以整合比较、种群和功能基因组数据，解决关于人类基因组中序列的功能和进化的基本问题。我们的软件和数据分析结果将免费提供给研究界。

DESCRIPTION (provided by applicant): Project Description Recent investments in data collection have produced rich catalogs of information about genomic function, human variation, and mammalian evolution, but improved computational tools are needed to integrate and interpret these catalogs, in order to permit the acquisition of new biological knowledge and advances in medicine. Here we outline an ambitious project to develop computational methods that will integrate publicly available data catalogs to provide deep new insights about the evolution and function of sequences in the human genome. Our proposal focuses in particular on noncoding sequences, which remain the most sparsely annotated and poorly understood regions of the genome. The proposal addresses three fundamental and closely related problems: (1) inference of human demography, to provide improved "null models" for statistical genetics and reveal local signatures of gene flow, natural selection, and other phenomena; (2) detection of natural selection on interspersed noncoding sequences such as transcription factor binding sites, to provide information about their function and the evolutionary processes that have shaped them; and (3) genome-wide prediction of "functional potential" based on integrated data sets, to identify new functional elements and prioritize candidate disease loci for experimental follow-up. Our proposal includes innovative statistical modeling, new algorithms for inference, the development of freely available software tools and browser resources, and detailed analyses of the latest genomic data sets. To our knowledge this will be the most comprehensive effort yet undertaken to integrate comparative, population, and functional genomic data in addressing fundamental questions about the function and evolution of sequences in the human genome. Our software and the results of our data analysis will be freely available to the research community.