Max Planck Institute for Evolutionary Biology, Plön, Germany.
PLoS Genet. 2012;8(8):e1002891. doi: 10.1371/journal.pgen.1002891. Epub 2012 Aug 30.
General parameters of selection, such as the frequency and strength of positive selection in natural populations or the role of introgression, are still insufficiently understood. The house mouse (Mus musculus) is a particularly well-suited model system to approach such questions, since it has a defined history of splits into subspecies and populations and since extensive genome information is available. We have used high-density single-nucleotide polymorphism (SNP) typing arrays to assess genomic patterns of positive selection and introgression of alleles in two natural populations of each of the subspecies M. m. domesticus and M. m. musculus. Applying different statistical procedures, we find a large number of regions subject to apparent selective sweeps, indicating frequent positive selection on rare alleles or novel mutations. Genes in the regions include well-studied imprinted loci (e.g. Plagl1/Zac1), homologues of human genes involved in adaptations (e.g. alpha-amylase genes) or in genetic diseases (e.g. Huntingtin and Parkin). Haplotype matching between the two subspecies reveals a large number of haplotypes that show patterns of introgression from specific populations of the respective other subspecies, with at least 10% of the genome being affected by partial or full introgression. Using neutral simulations for comparison, we find that the size and the fraction of introgressed haplotypes are not compatible with a pure migration or incomplete lineage sorting model. Hence, it appears that introgressed haplotypes can rise in frequency due to positive selection and thus can contribute to the adaptive genomic landscape of natural populations. Our data support the notion that natural genomes are subject to complex adaptive processes, including the introgression of haplotypes from other differentiated populations or species at a larger scale than previously assumed for animals. This implies that some of the admixture found in inbred strains of mice may also have a natural origin.
选择的一般参数,如自然种群中正向选择的频率和强度,或者基因渐渗的作用,仍然了解不足。家鼠(Mus musculus)是一个特别适合的模型系统,可以用来研究这些问题,因为它的亚种和种群有明确的分化历史,并且有广泛的基因组信息。我们使用高密度的单核苷酸多态性(SNP)分型阵列,评估了两个亚种 M. m. domesticus 和 M. m. musculus 的每个亚种的两个自然种群的基因组中正向选择和等位基因渐渗的模式。应用不同的统计程序,我们发现了大量受到明显选择作用的区域,表明稀有等位基因或新突变经常受到正向选择。这些区域中的基因包括经过深入研究的印迹基因座(例如 Plagl1/Zac1)、参与适应(例如α-淀粉酶基因)或遗传疾病(例如 Huntingtin 和 Parkin)的人类基因的同源基因。两个亚种之间的单倍型匹配揭示了大量来自特定亚种的特定种群的渐渗模式的单倍型,至少有 10%的基因组受到部分或完全渐渗的影响。通过中性模拟进行比较,我们发现渐渗单倍型的大小和比例与纯迁移或不完全谱系分选模型不兼容。因此,似乎渐渗单倍型可以由于正向选择而增加频率,从而有助于自然种群的适应性基因组景观。我们的数据支持这样一种观点,即自然基因组受到复杂的适应性过程的影响,包括来自其他分化种群或物种的渐渗单倍型,其规模比以前认为的动物要大。这意味着在近交系小鼠中发现的一些杂交可能也有自然起源。