CAREER: Next-generation inference of evolutionary paramaters from genome-wide sequence data

该项目的目标是开发一套新的统计和计算方法，用于从多个个体的全基因组推断详细的进化史。虽然测序技术的进步使DNA测序成为大多数实验室的例行公事，但从大型测序数据集推断进化历史的方法不得不做出限制性的或生物学上不切实际的假设，以实现计算上的易操作性。该项目将通过放宽这些限制性假设来克服目前方法的局限性，并将为用户提供评估这些方法所产生的估计的准确性的方法。这项研究将与旨在向高中年轻女性介绍计算机科学和生物学研究的多项活动组成的教育计划相结合：在PI的实验室提供暑期研究经验，邀请高中生与本科生合作，在PI的课程中向生物专业教授编程技能，并向各种高中观众教授编程。该项目的目标是开发准确推断(1)种群规模随时间的变化，(2)种群间迁移率随时间的变化，以及(3)自然选择的基因组目标的方法--所有这些都来自序列数据，取自同一物种中的多个个体。开发的方法将模拟重组，为报告的估计产生不确定性测量，允许复杂的种群历史，并识别正在选择的基因组区域。在这个项目中开发的方法也将被应用于测试关于一系列生物体进化史的假设。研究产生的出版物中产生的软件和分析的数据将在实验室的数据库(http://ramachandran-data.brown.edu/))上向公众提供，并通过存放在全面R档案网络(http://cran.r-project.org/).)上的R包向公众提供

The goal of this project is to develop a suite of novel statistical and computational methods for inferring detailed evolutionary histories from whole-genomes of multiple individuals. While advances in sequencing technology have made DNA sequencing routine in most laboratories, methods to infer evolutionary histories from large sequencing datasets have had to make restrictive or biologically unrealistic assumptions to achieve computational tractability. This project will overcome the limitations of current methods by relaxing such restrictive assumptions and it will provide users with ways to assess the accuracy of estimates produced by the methods. This research will be integrated with an education plan that consists of multiple activities intended to introduce young women in high school to computer science and biology research: offering summer research experiences in the PI's lab, inviting high school students to collaborate with undergraduates in the PI's courses that teach programming skills to biology majors, and teaching programming to a variety of high school audiences.The objectives of this project are to develop methods that accurately infer (1) changes in population size over time, (2) rates of migration between populations over time, and (3) genomic targets of natural selection --- all from sequence data alone, taken from multiple individuals within a single species. The methods developed will model recombination, produce measures of uncertainty for reported estimates, allow for complex population histories, and identify regions of the genome under selection. The methods developed in this project will also be applied to test hypotheses regarding the evolutionary histories of a range of organisms. Software produced and data analyzed in publications resulting from the research will be made available to the public on the lab's data repository (http://ramachandran-data.brown.edu/) and through R packages deposited on the Comprehensive R Archive Network (http://cran.r-project.org/).