Shi Cheng-Min, Liu Qi, Zhao Shilei, Chen Hua
CAS Key Laboratory of Genomic and Precision Medicine, Beijing Institute of Genomics, Chinese Academy of Sciences, Beijing, China.
University of Chinese Academy of Sciences, Beijing, China.
Ann Hum Genet. 2019 Sep;83(5):348-354. doi: 10.1111/ahg.12320. Epub 2019 Apr 26.
Ancestry informative markers play an important role in medical genetics and forensic analyses. Several ancestry informative SNP panels have been developed and validated that can differentiate global populations into continental or major regional groups. These global panels have served as good first-tier genetic markers; however, their performance in discriminating populations within regions appears unsatisfactory. To boost ancestry inference for regional populations, second-tier panels with more refined discrimination power among subpopulations within each of the regions need to be developed. In East Asia, Han Chinese, Japanese, and Korean show highly similar externally visible characteristics and are genetically closely related. Reliable ancestry informative genetic markers appear invaluable in discriminating these populations. In the present study, we compiled a genome-wide SNP dataset composing of 317,439 clean SNPs for a total of 1101 unrelated individuals from Han Chinese (817), Koreans (184), and Japanese (100). From this starting dataset, we developed a set of four nested ancestry informative SNP panels including 36, 59, 98, and 142 SNPs, respectively. The results of cross-validation tests indicate that these panels can discriminate the Chinese Han, Japanese, and Korean populations with overall average accuracies ranging from 90% to 99%. In the further performance assessments, these panels also manifested high sensitivity and specificity. In combination with the first-tier global panels, these second-tier panels would contribute to medical genetics and forensic research in East Asia.
祖先信息标记物在医学遗传学和法医分析中发挥着重要作用。已经开发并验证了几个祖先信息SNP面板，它们可以将全球人群区分为大陆或主要区域群体。这些全球面板已成为良好的一级遗传标记；然而，它们在区分区域内人群方面的表现似乎并不令人满意。为了提高对区域人群的祖先推断能力，需要开发在每个区域内的亚群体中具有更精细区分能力的二级面板。在东亚，汉族、日本人和韩国人表现出高度相似的外部可见特征，并且在基因上密切相关。可靠的祖先信息遗传标记物在区分这些人群方面似乎具有重要价值。在本研究中，我们为来自汉族（817人）、韩国人（184人）和日本人（100人）的总共1101名无关个体编制了一个由317,439个干净SNP组成的全基因组SNP数据集。从这个初始数据集出发，我们开发了一组四个嵌套的祖先信息SNP面板，分别包含36、59、98和142个SNP。交叉验证测试结果表明，这些面板能够区分中国汉族、日本人和韩国人群体，总体平均准确率在90%到99%之间。在进一步的性能评估中，这些面板也表现出高灵敏度和特异性。与一级全球面板相结合，这些二级面板将有助于东亚的医学遗传学和法医研究。
