Efficient Identification of Induced Mutations in Crop Species by Ultra-High Throughput DNA Sequencing

PI: Luca Comai（加州大学戴维斯分校）CoPIs: Jorge Dubcovsky（加州大学戴维斯分校）和Steven Henikoff （Fred Hutchinson癌症研究中心，西雅图）合作者：Robert Tran和Dawei Lin（加州大学戴维斯分校）关于基因序列和功能的信息不断增长，激发了人们对重要农艺和品质性状产生重大影响的途径的兴趣。然而，尽管越来越多的基因是有价值的失活靶点，但能够灵活靶向50 - 3000个作物基因的作物基因修饰方法有限。该项目将对TILLING （Targeting Induced Local lesion IN genomics）进行超高通量测序，这是一种发现感兴趣基因诱导突变的通用方法。利用水稻和小麦TILLING品系，它将利用聚合酶链反应（pcr）从汇集的基因组模板中介导扩增，以数十到数百个基因为目标。该研究将测试和实施几种技术和计算方法，以解决高度复杂模板中错误的真实突变。为了瞄准成百上千个基因，该项目将使用序列条形码技术，从成百上千个现有的拟南芥（Arabidopsis）以及后来的水稻和小麦的TILLING品系中，创建由短适配器连接的随机片段组成的DNA库。从扩增池中提取的Aliquots将使用定制的可编程阵列或连接到微阵列或微珠上的长寡核苷酸进行捕获程序，洗脱的片段将进行Solexa测序。该项目将建立一个经济有效的管道，作为用户支持的服务将是可持续的，取代更费力和昂贵的核苷酸错配检测技术。水稻和小麦跨越了关键特性的范围，因此可以作为所有作物的典范。它们在倍性、二倍体与异源多倍体、序列信息、高与低、基因组大小、小与大、突变密度、中等与非常高等方面存在差异。在开发TILLING的测序方法的同时，该项目将利用已经存在的和经过测试的种群来瞄准每种作物中的一组基因，这些基因将对科学家和育种者社区有用。因此，本研究将通过提供易于分离和分析的变异，促进改良作物品种的育种。本研究将为作物的高效耕作提供方法。通过这样做，它将刺激基础发现和有针对性的性状改进。例如，在小麦中，同源基因组的独立失活随后进行育种可以解决许多重要的农艺和品质性状。由于通过直接测序发现突变将允许研究人员从头开始选择要表征的突变，因此在计算机上收集诱导突变将为育种社区提供强大的资源。所建立的方法不仅可用于TILLING，还可用于发现自然变异，或鉴定数量性状位点的等位基因变化。连接外展计划结合了本科和专业培训。它基于三种方法：一个实习项目，将来自代表性不足人群的学生和教师与实验室到实地的基因组实验管道配对，一个数据库和方法网站，以及一个培训作物科学家的年度研讨会。数据和方法将通过项目网站（http://tilling.ucdavis.edu）和已建立的基因组数据库（http://www.gramene.org/; http://www.arabidopsis.org/）向公众开放。

PI: Luca Comai (University of California - Davis)CoPIs: Jorge Dubcovsky (University of California - Davis) and Steven Henikoff (Fred Hutchinson Cancer Research Center, Seattle)Collaborator: Robert Tran and Dawei Lin (University of California - Davis). The growing information on gene sequence and function stimulates interest in pathways expected to have large effects on important agronomic and quality traits. Nevertheless, while more and more genes are valuable inactivation targets, approaches for crop gene modification that can flexibly target 50 to 3000 genes per crop species are limited. This project will apply ultra-high throughput sequencing to TILLING (Targeting Induced Local Lesions IN Genomes), a general method for the discovery of induced mutations in genes of interest. Working with rice and wheat TILLING lines, it will employ PCR-mediated amplification from pooled genomic templates to target dozens to hundreds of genes. The research will test and implement several technical and computational methods to resolve real mutations from errors in highly complex templates. To target hundreds to thousands of genes, the project will use sequence barcoding to create DNA pools of short adapter-ligated random fragments from hundreds or thousands of existing TILLING lines of Arabidopsis, and later rice and wheat. Aliquots from an amplified pool will be subjected to a capture procedure using custom programmable arrays or long oligos attached to either microarrays or beads, and the eluted fragments subjected to Solexa sequencing. The project will build an economical and effective pipeline that will be sustainable as a user-supported service, superseding more laborious and expensive nucleotide mismatch-detection technology. Rice and wheat span the range of critical characteristics and can therefore serve as models for all crops. They differ in ploidy, diploid versus allopolyploid, sequence information, high versus low, genome size, small versus large, and mutation density, moderate versus very high. While developing a sequencing approach to TILLING, the project will use already existing and tested populations to target a set of genes in each crop that will be useful to the community of scientist and breeders. Thus, this research will facilitate breeding of improved crop varieties by providing facile isolation and analysis of variants. This research will provide methods for efficient TILLING of crop plants. By doing so, it will stimulate both basic discovery as well as targeted trait improvement. For example, in wheat independent inactivation of homeologous gene sets followed by breeding can address a host of important agronomic and quality traits. Because discovery of mutations by direct sequencing will allow researchers to choose ab initio which mutant to characterize, in silico collections of induced mutations will provide powerful resources for breeding communities. The developed methods will not only be useful for TILLING but also to discover natural variation, or to identify allelic changes underlying quantitative trait loci. The connected outreach program combines the undergraduate and professional training. It is based on three approaches: an internship program that will pair students and teachers from underrepresented populations with laboratory-to-field experimental genomic pipeline, a database and methods web site, and a yearly workshop for training crop scientists. Data and methods will be accessible to the public through the project web site (http://tilling.ucdavis.edu) and through established genomics databases (http://www.gramene.org/; http://www.arabidopsis.org/).