Origins and architecture of human genomic variation

我们的计划是关于识别，理解和解释人类群体中的遗传变异模式。进化史包含两个组成部分。首先是过去的人口统计学，可以用所讨论的基因组片段的谱系来表示;这些谱系是由迁移、创始人效应、遗传漂变、近亲繁殖、混合以及自然选择塑造的。其次是过去的遗传事件，突变和重组，在基因组记录中留下了标记。反过来，这些标记可以在样本个体携带的染色体中找到，作为标记来推断潜在的系谱，从而推断种群和基因组进化历史。我们要检验的假设主要来自我们以前的研究。我们建议解决三组问题，涉及（i）智人起源的时间，其在非洲的早期历史，非洲以外的扩张（s）和之后，集中我们的分析携带单倍型的DNA片段为古代混合物提供信息，（ii）鉴定携带古老混合物的推定特征的基因组区段（对于（i）项下的分析，通过寻找与参考古基因组共享衍生等位基因并且还受地理限制的SNP簇（例如，专门在非洲或非洲以外），和（iii）使用的差异签名的突变和重组记录的过去的人口事件。关于重组，我们使用两种方法，基于连锁不平衡系数-r2和四配子测试，如前所述。因为第一个“看到”只有交换（CO）和第二个基因转换（GC），我们建议研究两者的影响，这是非常新的：首先，在基因组规模上区分CO和GC的影响，其次，这也可以用来研究种群的历史。我们建立在我们早期的工作，包括调查的遗传多样性的X连锁区段在世界范围内的样本。这些研究使我们能够提出非正统的假设，关于非洲祖先人口的结构，非洲瓶颈的时间和地点，关于后来被证实是尼安德特人起源的非非洲遗传贡献，以及创始人效应在大陆，特别是美洲人口中的作用。为了解决这些问题，我们对我们收集的样本（n> 10，000（X染色体））进行了重新测序和基因分型，并从现有数据库中添加了信息。我们依靠我们的合作伙伴网络。我们开发了新的计算工具，以更好地从遗传记录中读取更多信息（例如时间创始人效应，研究适应性选择，估计重组率，寻找调节变体，或区分CO和GC等-参见CV）。目前，进化基因组学的专家越来越需要将实验室能力与统计遗传学的计算技能和知识相结合。通过桥接不同的学科，我们的研究计划促进跨学科的举措，并成为在基础和应用科学，如生物信息学和人口基因组学培训的新一代专家的温床。为了填补这些领域的教学空白，我在2006年开始了一项教学计划，从那时起一直作为蒙特利尔春季学校的年度版本。我们研究计划的基因组学（计算和实验）和群体遗传学方向为未来的进化生物学学者提供了一个良好的培训环境。这是尽管我们专注于我们物种的历史，它的人口和遗传学，有趣的本身，但方法，方法和应用程序是普遍使用的。

Our program is about recognizing, understanding and explaining patterns of genetic variation among human populations. Evolutionary history involves two components. First is past demography that can be represented by genealogies of genomic segments in question; these genealogies were sculptured by migration, founder effects, genetic drift, inbreeding, admixture as well as natural selection. Second are past genetic events, mutations and recombinations that left marks in the genomic record. In turn, these marks, which can be found in the chromosomes carried by sampled individuals, serve as markers to infer the underlying genealogies and thus the populations and genomic evolutionary history. The hypotheses we want to examine largely grew from our previous studies. We propose to address three groups of questions concerning (i)the time of the origin of H.sapiens, its early history in Africa, the out of Africa expansion(s) and after, focusing our analyses on DNA segments carrying haplotypes informative for archaic admixture, (ii) the identification of genomic segments carrying putative signatures of archaic admixture (for analyses under (i) by looking for clusters of SNPs that share derived alleles with a reference archaic genome and are also geographically restricted (e.g. exclusively in or outside Africa), and (iii) the use of differential signatures of past demographic events recorded by mutation and by recombination. Concerning recombination, we use two methods, based on linkage disequilibrium coefficient-r2 and on four-gamete test, described earlier. Because the first “sees” only crossing over (CO)and the second gene conversion (GC) as well, we propose to study the effect of both, and this is very new: first, to distinguish between the effects of CO and GC at the genomic scale, and second, how this can be also used to study populations’ history. We build on our earlier work that includes investigations of genetic diversity of X-linked segments in world-wide sample. These studies allowed us to forward unorthodox hypotheses about the structured ancestral African population, the time and place of the out-of-Africa bottleneck, about the non-African genetic contribution subsequently confirmed to be of the Neanderthal origin and the role of founder effects in the peopling of continents and of the Americas in particular. To pursue these questions, we use resequencing and genotyping of samples from our collection (n>10,000 (of X chromosomes), adding information from available databases. We rely on our extended network of collaborators. We develop new computational tools to read better and more from the genetic record (such as to time founder effects, study adaptive selection, to estimate recombination rate, finding regulatory variants, or distinguishing between CO and GC, etc. –see CV). Currently, there is a growing need of specialists in evolutionary genomics integrating laboratory competence with computational skills and knowledge in statistical genetics. By bridging different disciplines, our research program promotes interdisciplinary initiatives and becomes a breeding ground of new generation of specialists trained in basic and applied sciences such as bioinformatics and population genomics. To fill a gap in teaching in these fields, in 2006 I started a teaching initiative, which continues since then as yearly editions of Montreal Spring School. The genomics (computational and experimental) and population genetics orientation of our research program provides an excellent training environment for future scholars in evolutionary biology at large. This is in spite of the fact that we focus on the history of our species, its populations and genetics, interesting by itself, but the methods, approaches and applications are of general use.