遗传漂变和自然选择是微进化过程中的两个核心驱动力。丰富的全基因组遗传多态数据为阐明这两个因素在微进化中的作用机制提供了前所未有的契机。另一方面，现有的群体遗传学理论和方法在个体基因组数据分析中具有计算强度大、稳健性差等局限性。本项目围绕以上问题，进行如下的新方法研究。首先，我们在递次式马尔可夫溯祖理论的框架上，提出新的近似模型和解析形式，构建能对来自多个群体的多个个体基因组数据进行高效分析的方法，以解决现有方法稳健性低或计算强度大的缺点；其次，我们针对目前软性自然选择研究难度大、检测方法缺乏的需求，提出了一个有效的统计方法。以上工作将为基因组时代研究微进化提供有效的数据分析工具，为进一步深入解析微进化的作用机制奠定基础。

Genetic drift and natural selection are essential driving forces in microevolution. The advent of genomic polymorphism data provides unprecedented opportunity for addressing their mechanism in microevolution. On the other hand, existing theory and methods have limitations when applying to personal genomic data analysis. This project develops new statistical methods aiming to solving these issues. First, we develop new approximation and analytical theory in the framework of sequential Markovian coalescent, and construct new methods for inferring demography of multiple populations using multiple personal genomes, for which existing methods are facing intensive computation or lack of robustness. Second, we propose a novel statistical method for detecting soft selective sweeps, which is of fundamental importance while without efficient detecting methods so far. The proposed research will provide efficient computational methods for analyzing genomic data and help elucidate the mechanism of microevolution.