A TWO-PRONGED STRATEGY TO GENERATE FULL-LENGTH CDNA'S

这项提议旨在产生可广泛用于全长测序程序的经验证的非冗余排列的全长cDNA集。实现这一目标有赖于实施一项双管齐下的战略，利用互补的资源和技术。我们计划中的湿实验室和信息学研究部分是完全集成的，它们对于成功实现我们的目标同样重要。简而言之，我们建议将5‘端富集的随机启动文库与寡聚[DT]启动的全长富集库结合使用，以产生杂交选择的文库，这些文库将大大丰富全长cDNA。对从这些混合选择的文库中产生的5‘EST数据的信息学分析将使cDNA能够分为三类：全长、全编码和不完全。基于这些分析，全长cDNA将被识别和重新排列。将产生三组这样重新排列的cDNA。第一个将包括从寡聚[DT]启动的全长富集库中挑选出的15,000个全长和全编码混合cDNA的集合。第二和第三组相对于第一组将是非冗余的，每组都将包含7,500个cDNA，它们来自一对寡聚[DT]启动的全长富集型和5‘端富集型随机引物库。加在一起，后两组将允许额外的15,000个mRNA的全长表达。我们的策略的核心是利用相同大小的分级细胞质mRNA来构建全长富集库和5‘端富集库。这对文库将为从2.37 kb到9.4kb的四个人mRNA大小片段中的每一个生成。将使用两种不同的方法来构建5‘端丰富的随机启动文库：寡核苷酸封顶技术和一种新的方法-“末端拯救”-它们的开发构成了本申请中提出的工作的一部分。对从这两种方法产生的文库中获得的5‘EST数据的信息学分析将有助于识别最佳的5’端富集文库，这些文库将被选择用于我们提出的双管齐下的全长cDNA阵列集的生成策略。

This proposal is aimed at the generation of verified non-redundant arrayed sets of full-length cDNAs to be made widely available for full-length sequencing programs. The achievement of this goal relies on the implementation of a two-pronged strategy that takes advantage of complementary resources and technologies. The wet laboratory and informatics research components of our proposal are totally integrated and they are both equally important for the successful achievement of our goals. Briefly, we propose to use 5' end-enriched random primed libraries in conjunction with oligo-[dT] primed full-length- enriched libraries to generate hybrid-selected libraries that will be greatly enriched for full-length cDNAs. Informatics analysis of 5' EST data generated from these hybrid-selected libraries will enable classification of cDNAs into three groups: full-length, full-coding and incomplete. Based on these analyses, full-length cDNAs will be identified and re-arrayed. Three such re-arrayed sets of cDNAs will be produced. The first will comprise a collection of 15,000 full-length and full-coding hybrid selected cDNAs from the oligo-[dT] primed full-length enriched libraries. The second and third sets, which will be non-redundant respective to the first set, will each contain 7,500 cDNAs, derived from a pair of oligo-[dT] primed full-length enriched and 5' end-enriched random primed libraries. Together, the latter two sets will allow for full-length representation of an additional 15,000 mRNAs. Central to our strategy is the utilization of the same size fractionated cytoplasmic mRNA for construction of both full-length-enriched and 5' end-enriched libraries. This pair of libraries will be generated for each of four human mRNA size fractions, spanning from 2.37 to 9.4 kb. Two different approaches will be used for construction of 5' end-enriched random primed libraries: the oligo-capping technology, and a novel method - "end rescuing" - whose development constitutes part of the work proposed in this application. Informatics analysis of 5' EST data obtained from libraries generated by both methods will be instrumental for identification of the best 5' end-enriched libraries to be selected for use in our proposed two-pronged strategy for generation of arrayed sets of full-length cDNAs.