Population genomics of the selective effects of new mutations

新突变选择性效应的群体基因组学

• 批准号: 9340237
• 负责人: Kirk Lohmueller
• 金额: $ 30.42万
• 依托单位: UNIVERSITY OF CALIFORNIA LOS ANGELES
• 依托单位国家: 美国
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-09-01 至 2021-05-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9340237
• 关键词: Address; Affect; Amino Acids; Biological; Biological Process; Canis familiaris; Complex; Computing Methodologies; Data; Data Set; Disease; Environment; Evolution; Future; Genetic; Genetic Epistasis; Genetic Polymorphism; Genetic Variation; Genetic study; Genome; Genomics; Heart; Homo; Human; Human Genome; Individual; Knowledge; Medical Genetics; Methods; Modeling; Mutation; Natural Selections; Nature; Pathogenicity; Population; Population Genetics; Public Health; Research; Risk; Running; Sampling; Site; Sum; Testing; Variant; Wolves; Work; cohort; computerized tools; exome sequencing; fitness; genetic pedigree; genetic variant; improved; next generation sequencing; novel; pressure; public health relevance; response; trait; transmission process

PROJECT SUMMARY (DESCRIPTION) How new mutations affect an individual's fitness lies at the heart of many questions in evolutionary genetics and is highly relevant for elucidating the genetic basis of complex traits. Current evidence suggests that while many mutations are deleterious, selection coefficients for particular mutations come from a distribution of se- lective effects (DSE). Although the DSE has been subjected to intense study over the past decade, many open questions about this distribution and the importance of deleterious mutations in evolution remain unanswered. The widespread use of next-generation sequencing has provided a plethora of new polymorphism datasets that are potentially informative about the nature of fitness effects of mutations. However, existing computational approaches make simplifying assumptions and lack statistical power to extract the full potential of these data. This project will develop and apply new computational approaches to integrate these emerging polymorphism datasets from multiple species together in a rigorous statistical framework to estimate fundamental parameters relating to natural selection across genomes. We will develop novel computational methods that use very large samples of human genomes to better estimates selective effects. First, we propose to leverage runs of homo- zygosity within an individual to co-estimates dominance coefficients and the DSE. We will develop a second set of methods that utilizes transmission patters within large numbers of pedigrees to estimate selective ef- fects. Third, previous studies of the DSE have assumed all sites are independent of each other and have ig- nored epistasis. We will develop a composite likelihood approach to estimate the degree of epistasis between many pairs of SNPs. We will then apply these methods to the large cohorts of human genomes that are being sequenced to obtain the most detailed estimates of the DSE, dominance coefficients, and nature of epistasis from human populations. Next, these new computational tools will be combined with data from a variety of dif- ferent species to address longstanding questions concerning selective effects. We will investigate the evolution of the DSE itself by testing whether it has changed in response to recent shifts in the environment. As an ex- ample, we will test whether the change in selective pressure associated with dog domestication has led to dif- ferences in the DSE between dogs and wolves. This will be accomplished by generating exome sequencing data from of dogs and wolves. Finally, we will investigate why different species appear to show disparate amounts of amino acid adaptation. This work will resolve a current paradox in population genetics and will con- tribute to understanding the different modes of adaptation across different species. In sum, successful comple- tion of this research will provide the most comprehensive estimates of dominance, the DSE, and the amount of epistasis in natural populations to date. These estimates will dramatically improve the interpretation of genetic variation in future evolutionary and medical genetic studies.

项目概要（描述） 新的突变如何影响个体的适应性是进化遗传学中许多问题的核心 并且与阐明复杂性状的遗传基础高度相关。目前的证据表明，虽然 许多突变是有害的，特定突变的选择系数来自于Se的分布， 选择性效应（DSE）。尽管DSE在过去十年中受到了深入的研究，但许多开放的 关于这种分布和有害突变在进化中的重要性的问题仍然没有答案。 下一代测序的广泛使用提供了大量新的多态性数据集 这是潜在的信息性质的健身效果的突变。然而，现有的计算 这些方法简化了假设，缺乏统计能力来充分挖掘这些数据的潜力。 本计画将发展并应用新的计算方法来整合这些新兴的多态现象 在严格的统计框架中将多个物种的数据集结合在一起，以估计基本参数 关于基因组间的自然选择。我们将开发新的计算方法， 人类基因组样本，以更好地估计选择性效应。首先，我们建议利用同性恋的运行- 个体内的接合性来共同估计显性系数和DSE。我们将开发第二个 一套方法，利用大量的谱系内的传输模式，以估计选择性的影响， 感染。第三，以前的DSE研究假设所有的网站是相互独立的，并有IG- 非上位性。我们将开发一种复合似然方法来估计 许多对SNPs。然后，我们将把这些方法应用于正在进行的人类基因组的大型队列。 排序以获得DSE、优势系数和上位性性质的最详细估计 从人群中。接下来，这些新的计算工具将与来自各种不同的数据相结合， 物种来解决长期存在的关于选择性效应的问题。我们将研究 通过测试DSE本身是否已根据环境中最近的变化而发生变化。作为一个前- 例如，我们将测试与狗驯化相关的选择压力的变化是否导致了差异， 在DSE中狗和狼之间的推论。这将通过生成外显子组测序来完成 数据来自狗和狼。最后，我们将研究为什么不同的物种似乎表现出不同的 氨基酸的适应性。这项工作将解决目前在群体遗传学的悖论，并将继续- 这有助于理解不同物种之间的不同适应模式。总之，成功的完成了- 这项研究的结果将提供最全面的估计优势，DSE，和数量， 自然种群中的上位性。这些估计将极大地改善对遗传学的解释。 在未来的进化和医学遗传学研究的变化。