A Statistical Method for Scanning the Genome for Regions with Rare Disease Alleles
A Statistical Method for Scanning the Genome for Regions with Rare Disease Alleles
复制标题
DOI：
10.1002/gepi.20483
复制
发表时间：
2010-07-01
影响因子：
2.1
通讯作者：
Garner, Chad
中科院分区：
文献类型：
--
作者：
Garner, Chad
确定大量个体的DNA序列是一项不切实际的任务，阻碍了研究稀有等位基因在常见疾病中的作用。下一代DNA测序技术正在开发中，这将使常见疾病的遗传学研究有可能研究遗传变异的全频谱，包括罕见的等位基因。这份报告描述了一种扫描基因组以寻找疾病易感区域的方法，该方法显示在疾病病例样本中与种族匹配的对照样本相比，稀有等位基因的数量增加。该方法基于隐马尔可夫模型，并通过似然比统计量来衡量罕见等位基因特征的疾病易感区域的统计支持度。由于缺乏经验数据，对该方法进行了仿真评估。在序列生成和隐马尔可夫模型参数范围内，分别在零假设和替代假设下测试了该方法的性能。结果表明，统计方法在识别真正的疾病易感区域方面表现良好，并且其性能主要受中性序列的变异量和在疾病易感区域中发现的罕见疾病等位基因的数量的影响。吉内。埃皮米诺。34：386-395,2010。(C)2010年Wiley-Liss公司
Studying the role of rare alleles in common disease has been prevented by the impractical task of determining the DNA sequence of large numbers of individuals. Next-generation DNA sequencing technologies are being developed that will make it possible for genetic studies of common disease to study the full frequency spectrum of genetic variation, including rare alleles. This report describes a method for scanning the genome for disease susceptibility regions that show an increased number of rare alleles among a sample of disease cases versus an ethnically matched sample of controls. The method was based on a hidden Markov model and the statistical support for a disease susceptibility region characterized by rare alleles was measured by a likelihood ratio statistic. Due to the lack of empirical data, the method was evaluated through simulation. The performance of the method was tested under the null and alternative hypotheses under a range of sequence generating and hidden Markov models parameters. The results showed that the statistical method performs well at identifying true disease susceptibility regions and that performance was primarily affected by the amount of variation in the neutral sequence and the number of rare disease alleles found in the disease susceptibility region. Genet. Epidemiol. 34 : 386-395, 2010. (C) 2010 Wiley-Liss, Inc.