标题： 压缩基因分型 显示英文标题

标题： Compressed Genotyping

摘要： 近年来，大量知识被积累起来，将微妙的遗传变异与从囊性纤维化到智力迟钝的各种医学疾病联系起来。 然而，由于DNA测序过程相对繁琐且昂贵，将这些知识常规应用于临床仍面临巨大挑战。 由于导致这些疾病的遗传多态性在人类群体中相对罕见，因此疾病相关多态性的存在或缺失可以被视为一种稀疏信号。 利用压缩感知和组测试的方法和思路，我们开发了一种成本效益高的基因分型协议。 特别是，我们将方案适应于最近开发的一类高通量DNA测序技术，并构建了一个数学框架，该框架与“传统”压缩感知理念有一些重要的区别，以应对不同的生物和技术约束。 显示英文摘要

摘要： Significant volumes of knowledge have been accumulated in recent years linking subtle genetic variations to a wide variety of medical disorders from Cystic Fibrosis to mental retardation. Nevertheless, there are still great challenges in applying this knowledge routinely in the clinic, largely due to the relatively tedious and expensive process of DNA sequencing. Since the genetic polymorphisms that underlie these disorders are relatively rare in the human population, the presence or absence of a disease-linked polymorphism can be thought of as a sparse signal. Using methods and ideas from compressed sensing and group testing, we have developed a cost-effective genotyping protocol. In particular, we have adapted our scheme to a recently developed class of high throughput DNA sequencing technologies, and assembled a mathematical framework that has some important distinctions from 'traditional' compressed sensing ideas in order to address different biological and technical constraints.