全球男性人口变迁历史的遗传统计学研究

发布时间：2018-06-24 08:21

本文选题：HGDP-CEPH + Y-SNP　；参考：《中国协和医科大学》2009年博士论文

【摘要】：关于现代人类的起源一直存在着争议,但目前绝大多数学者接受“走出非洲”学说,即认为:十万年前现代人起源于非洲的某个较小人群,随着人口数量的不断增长,逐步迁移到非洲以外的很多地方。在过去近三十年里,遗传学不断地渗透到人类起源的研究中,并且提供了众多支持“走出非洲”学说的DNA分子水平证据。在此基础上,有学者进一步提出了“系列建立者模型”学说:当撒哈拉以南非洲某个群体达到一定人口饱和度时,其中的一部分人就会迁移到远离其原居住地的另外一个地区建立一个新群体,当新群体的人口再次增长到较高的饱和度时,就会发生第二次迁移,这个过程反复不断,直至达到地球上最远的人类聚居点——美洲大陆。本研究应用遗传学数据和遗传统计学原理系统地探讨了这个过程中的人口变迁历史,如世界不同地区群体的最近共同祖先年代(TMRCA)、有效群体的大小(Ne)、人口数量增长的起始时间以及人口的增长率等问题。本研究应用BATWING程序系统地分析了HGDP-CEPH(人类基因组多样性计划-人类基因组多样性中心)样本库中代表全球绝大部分现代人群的51个群体,其中包括590例男性的37个Y-SNPs和65个Y-STRs基因型数据。BATWING程序根据输入的遗传学数据和先验参数(prior),用基于溯祖理论的马尔科夫链蒙特卡罗算法模拟一系列(如十万个)谱系树(genealogical tree),同时计算符合每个谱系树结构的人口学后验参数(posterior),即TMRCA、Ne、人口数量增长的起始时间以及人口的增长率。在程序模拟达到平衡后,可以得出每个参数的后验估计及其可信区间。在方法学上,本研究首先系统地分析了不同类型(简单和复杂)的Y-STR、不同数量的Y-STR、样本量的大小以及不同的STR突变率对人口学参数后验分布产生的影响。结果显示:(1)不同类型的Y-STR对人口学参数的估计基本没有影响。应用较多的STR相对应用较少STR具有一定的优势,可以推算出更准确的TMRCA,所以在分析不同群体的人口学参数时,应采用全部65个Y-STR;(2)样本量为4个个体时(本研究所采用的最小的样本量)即可提供有价值的人口学参数信息,不需要剔除或合并这样的群体;(3)将三种STR突变率(rEMR,EMR和OMR)分别应用于对51个群体的TMRCA、人口数量增长的起始时间、Ne以及人口增长率的估算,发现前三个人口学参数后验估计的大小顺序为rEMR＞EMR＞OMR,但人口增长率的大小顺序与之相反。由于应用rEMR对人口学参数的估计更接近于考古学上的估计(最可信的估计),且rEMR根据变异度将OMR和EMR进行了很好的组合,因而我们最终采用了rEMR。在人口学上,我们观察到了一个普遍的模式:撒哈拉沙漠以南非洲人群的TMRCA最古老、Ne最大且人口数量增长的起始时间最早,而美洲人群的TMRCA最年轻、Ne最小且人口数量增长的起始时间最晚,其他大陆人群的这些参数介于二者之间,形成一个梯度。所有非洲以外人群的TMRCA、Ne及人口数量增长的起始时间与该人群居住地到东非间的步行距离呈显著的负相关,这与大多数关于人类遗传变异及进化史的研究结果相符;与之相反,人口增长率与距离的远近呈正相关,但相关性很弱;Y染色体谱系中不同分支的多样性及其TMRCA也随着与东非之间距离的增加而降低。所有这些结果均支持“系列建立者模型”学说。同时,我们还发现了一些群体呈现特殊模式,以往的研究中也曾观察到其中的某些特例:如非洲Yoruba人群、欧洲Basque人群和北亚Yakut人群,他们的人口数量增长的起始时间明显晚于该地区的其他人群;但是本研究还发现了几个其它特例群体,例如巴基斯坦Kalash和Hazara的人口增长率较低,北非Mozabite人口数量增长的起始时间很晚。本研究利用遗传学数据,应用遗传统计学的策略提供了迄今为止最完整、最详细的全球男性人口变迁历史的信息。为了估测如此复杂的人类变迁过程,我们不得不简化某些分析细节,比如在BATWING模拟中,我们只假设每个人群的初始有效群体大小不变,经过一段时间后开始指数扩增,并未考虑更复杂的人口变化、群体之间的融合以及其他一些特殊的历史事件。然而,本研究却有力地阐明了人类变迁历史中最关键的特征。无论是全球人口学变迁的宏观模式,还是某些微观特例,本研究的结论都与其他大多数遗传学和非遗传学研究的结果相吻合。同时,还有一小部分特殊群体为我们今后的研究提供了新的方向。
[Abstract]:There has been a dispute about the origin of modern human beings, but the vast majority of scholars now accept the "go out of Africa" theory that one hundred thousand years ago, modern people originated from a small population in Africa and gradually moved to many places outside Africa as the population grew, and in the past thirty years, genetics continued to infiltrate. On the basis of the study of human origin and the number of DNA molecular evidence supporting the "out of Africa" theory, some scholars have further proposed the "series builder model" theory: when a group of sub Saharan Africa reaches a certain population saturation, some of them will move away from their original origin. Another area in the place of residence establishes a new group, and when the population of the new population grows again to a higher saturation, second migration will occur, and the process is repeated until it reaches the most distant human settlement point on the earth - the American continent. The history of population changes in this process, such as the recent common ancestor age (TMRCA) of different regions of the world, the size of the effective population (Ne), the beginning of the population growth and the rate of population growth.
In this study, the BATWING program was used to systematically analyze 51 groups representing the vast majority of the world's modern population in the HGDP-CEPH (human genome diversity program - human genome diversity center) sample library, including 37 Y-SNPs and 65 Y-STRs genotype data.BATWING programs for men, based on the input genetic data and first. Prior, the Markov chain Monte Carlo algorithm based on abduction theory is used to simulate a series of (such as the one hundred thousand) pedigree tree (genealogical tree), at the same time, to calculate the demographic posterior parameter (posterior), that is, TMRCA, Ne, population growth and population growth rate that conforms to the structure of each pedigree. After equilibrium, a posteriori estimation of each parameter and its confidence interval can be obtained.
In methodology, this study first systematically analyzed the effects of different types (simple and complex) Y-STR, the number of Y-STR, the size of the sample size and the different STR mutation rates on the posterior distribution of demographic parameters. The results showed that: (1) different types of Y-STR had no effect on the estimation of demographic parameters. More STR was used. Relatively less application STR has a certain advantage to calculate a more accurate TMRCA, so in the analysis of the demographic parameters of different groups, all 65 Y-STR should be used; (2) the sample size of 4 individuals (the minimum sample size used in this study) can provide valuable value of the demographic information, do not need to eliminate or merge such (3) (3) three kinds of STR mutation rates (rEMR, EMR and OMR) were applied to the TMRCA of 51 populations, the initial time of population growth, Ne and the estimation of population growth rate, and the order of the first three oral parameters was found to be rEMR > EMR > OMR, but the order of the population growth rate was the opposite. The application of rEMR to people was the opposite. The estimation of the oral parameters is closer to the archaeological estimate (the most credible estimate), and rEMR makes a good combination of OMR and EMR based on the variation, so we finally use the rEMR.
In demographics, we have observed a universal model: the oldest TMRCA in the sub Saharan African population, the earliest beginning of the largest and population growth of Ne, and the youngest population of the Americas, the smallest Ne and the earliest population growth, and these parameters between the other continent population are between two. A gradient. The start time of TMRCA, Ne and population growth of all people outside Africa was significantly negatively correlated with the walking distance between the population and the East Africa, which was in line with most studies on human genetic variation and evolutionary history; in contrast, the growth rate of human mouth was positively related to distance distance, but related to the distance. The diversity of the different branches of the Y chromosome lineage and its TMRCA also decreased with the increase of the distance from East Africa. All these results support the "series builder model" theory. At the same time, we also found that some groups present special patterns, and some special cases in previous studies, such as African Yorub, have also been observed. The population growth of the a population, the European Basque population and the North Asian Yakut population was significantly later than the rest of the population in the region, but the study also found that several other special groups, such as the Pakistan Kalash and Hazara, had a lower population growth rate and a late start of the population growth of the North African Mozabite population.
This study uses genetic data and the strategy of applying genetic statistics to provide the most complete and detailed information about the history of global male population change so far. In order to estimate such a complex process of human change, we have to simplify some of the detailed analysis, for example, in the BATWING simulation, we only assume the initial effectiveness of each group. The size of the population is constant, and the exponential expansion after a period of time does not take into account more complex demographic changes, group integration and other special historical events. However, this study forcefully illustrates the most critical characteristics in the history of human changes. In particular, the findings of this study are consistent with the results of most other genetic and non genetic studies. At the same time, a small number of special groups provide a new direction for our future research.
【学位授予单位】：中国协和医科大学
【学位级别】：博士
【学位授予年份】：2009
【分类号】：R394

【共引文献】