Dutta Rajib, Mainsah Joseph, Yatskiv Yuriy, Chakrabortty Sharmistha, Brennan Patrick, Khuder Basil, Qiu Shuhao, Fedorova Larisa, Fedorov Alexei
Program in Biomedical Sciences, University of Toledo, Health Science Campus, Toledo, 43614, OH, USA.
Department of Medicine, University of Toledo, Health Science Campus, Toledo, 43614, OH, USA.
BMC Genomics. 2017 Jun 5;18(1):433. doi: 10.1186/s12864-017-3776-5.
Inferring history from genomic sequences is challenging and problematic because chromosomes are mosaics of thousands of small Identicalby-descent (IBD) fragments, each of them having their own unique story. However, the main events in recent evolution might be deciphered from comparative analysis of numerous loci. A paradox of why humans, whose effective population size is only 10, have nearly three million frequent SNPs is formulated and examined.
We studied 5398 loci evenly covering all human autosomes. Common haplotypes built from frequent SNPs that are present in people from various populations have been examined. We demonstrated highly non-random arrangement of alleles in common haplotypes. Abundance of mutually exclusive pairs of common haplotypes that have different alleles at every polymorphic position (so-called Yin/Yang haplotypes) was found in 56% of loci. A novel widely spread category of common haplotypes named Mosaic has been described. Mosaic consists of numerous pieces of Yin/Yang haplotypes and represents an ancestral stage of one of them. Scenarios of possible appearance of large number of frequent human SNPs and their habitual arrangement in Yin/Yang common haplotypes have been evaluated with an advanced genomic simulation algorithm.
Computer modeling demonstrated that the observed arrangement of 2.9 million frequent SNPs could not originate from a sole stand-alone population. A "Great Admixture" event has been proposed that can explain peculiarities with frequent SNP distributions. This Great Admixture presumably occurred 100-300 thousand years ago between two ancestral populations that had been separated from each other about a million years ago. Our programs and algorithms can be applied to other species to perform evolutionary and comparative genomics.
从基因组序列推断历史具有挑战性且存在问题，因为染色体是由数千个小的同源片段（IBD）拼接而成，每个片段都有其独特的故事。然而，近期进化中的主要事件或许可通过对众多基因座的比较分析来解读。一个关于有效种群规模仅为10的人类为何拥有近三百万个常见单核苷酸多态性（SNP）的悖论被提出并加以研究。
我们研究了均匀覆盖所有人类常染色体的5398个基因座。对由不同人群中存在的常见SNP构建的常见单倍型进行了研究。我们证明了常见单倍型中等位基因的高度非随机排列。在56%的基因座中发现了大量在每个多态性位置具有不同等位基因的互斥常见单倍型对（即所谓的阴阳单倍型）。描述了一种新的广泛分布的常见单倍型类别——镶嵌型。镶嵌型由众多阴阳单倍型片段组成，代表了其中一个的祖先阶段。使用先进的基因组模拟算法评估了大量常见人类SNP可能出现的情况及其在阴阳常见单倍型中的惯常排列。
计算机建模表明，观察到的290万个常见SNP的排列不可能源自单一孤立种群。提出了一个“大混合”事件，它可以解释常见SNP分布的特殊性。这个大混合大概发生在10万至30万年前，是在大约100万年前彼此分离的两个祖先种群之间。我们的程序和算法可应用于其他物种以进行进化和比较基因组学研究。
