Elucidating and detecting adaptive introgression

在进化过程中，人类已经适应了许多具有挑战性的环境。例如，不同的温度、饮食、病原体和海拔导致了当地的适应。适应性是通过种群中有益基因突变的选择和增加而发生的，这些基因突变可以通过随机突变自发产生，也可以通过与另一个种群的混合而获得。对人类基因组的分析表明，我们的一些有益突变来自尼安德特人和丹尼索瓦人等古人类群体，数万年前现代人与这些群体的杂交促进了这些突变的发生。这个过程被称为自适应渗透。虽然有一些由适应性基因渗入产生的有益基因突变的例子，但在本项目中，将建立工具来扫描人类基因组，以确定更多的适应性基因渗入候选基因，并充分表征这一过程在许多人类群体中的重要性。重要的是，该项目将包括对本科生和研究生进行计算科学和基因组学方面的培训，并且创建的工具将免费供其他人使用。此外，将为本科生开设一门新的跨学科课程，将人类遗传变异数据分析、编程、统计学和生物学联系起来。适应性的检测和表征主要通过两种模型来实现：新生突变选择（SDN）和常住变异选择（SSV）。因此，假设一个种群要么必须等待一个有益的突变从头出现，要么它拥有足够的中立变异，可以在环境变化下变得有益。然而，大多数种群并不是孤立生活的，它们通过混合（基因从供体种群流向受体种群）与其他种群交换了遗传变异。这个过程是一种进化的力量，可能会加速接受者群体的适应。PI将共同建立正选择和基因流模型，以研究该模型下的遗传变异模式，与其他两种正选择模型（SDN， SSV）进行比较和对比，确定哪些数据摘要能够准确区分适应性基因流与SDN和SSV，开发准确检测这种选择类型的新统计数据，并开发统计和计算工具来扫描基因组以识别候选区域。这些工具将应用于人类和其他生物体的真实数据集。更广泛的影响包括培养博士后和学生，建立一个以项目为导向的课程，整合多个学科（编程，统计和建模），教学生可视化生物数据集进行探索性分析，检验假设并将模型拟合到数据中，并通过一系列讲座将科学带给高中生。最后，在这项资助下开发的所有计算工具将免费提供给科学界。

Humans have adapted to many challenging environments during our evolution. For example, different temperatures, diets, pathogens and altitudes have led to local adaptations. Adaptations occur through the selection and increase of beneficial mutations in genes in a population, which can arise spontaneously by random mutation or can be acquired through admixture with another population. Analysis of human genomes has shown that some of our beneficial mutations have come from archaic human populations like the Neanderthals and Denisovans, facilitated by interbreeding between modern humans and those groups tens of thousands of years ago. This process is referred to as adaptive introgression. While there are a few examples of beneficial mutations in genes arising from adaptive introgression, in this project tools will be built to scan human genomes to identify more candidate genes for adaptive introgression and to fully characterize the importance of this process in many human populations. Importantly, the project will involve training of undergraduate and graduate students in computational science and genomics, and the tools created will be freely accessible for others to use. In addition, a new interdisciplinary course that bridges data analysis of human genetic variation, programming, statistics and biology will be offered to undergraduate students. Detecting and characterizing adaptation has mostly been approached through two models: selection on de novo mutations (SDN) or selection on standing variation (SSV). Therefore it is assumed that a population either has to wait for a beneficial mutation to arise de novo or it harbors enough neutral standing variation that can become beneficial under a change in environment. However, most populations do not live in isolation and have exchanged genetic variants with other populations through admixture (gene-flow from a donor population to a recipient population). This process is an evolutionary force that may accelerate adaptation in the recipient population. The PI will model positive selection and gene-flow jointly to investigate the patterns of genetic variation under this model, to compare and contrast to the two other models of positive selection (SDN, SSV), to determine what summaries of the data accurately distinguishes adaptive gene flow from SDN and SSV, to develop novel statistics that accurately detect this type of selection and to develop statistical and computational tools to scan genomes to identify candidate regions. These tools will be applied to real data sets in humans and in other organisms. The broader impacts include training a postdoc, students, building a project-oriented course that integrates multiple disciplines (programming, statistics and modeling) that will teach students to visualize biological data sets for exploratory analyses, to test hypothesis and to fit models to the data, and bringing science to high school students through a series of lectures. Finally, all computational tools developed under this grant will be made freely available to the scientific community.