PARALLEL DNA SEQUENCE ASSEMBLY PROGRAM

描述（改编自研究者摘要）：研究者 将开发一种软件工具，用于组装 大规模的鸟枪测序项目 该软件将进行测试 第一个是由计算机从大规模DNA序列中产生的DNA片段 然后是大规模测序项目中的真实的DNA片段。 该软件将免费分发给非营利组织。 的 研究者将协助将软件集成到测序中 基因组中心的环境。 本项目的目标将通过以下两个主要方面来实现 对以前开发的DNA序列组装程序的改进。 第一个改进是制定解决问题的战略 由重复序列引起的。 在这个策略中，所有的碎片 从一个重复的序列被识别出来， 使用有关的附加信息来解决片段的组装 位于重复序列拷贝侧翼的片段。 第二个改进是增加了装配的能力 通过在PVM中开发程序的并行版本来实现程序 本地计算机网络上的并行编程环境。 的 调查员将把调查中两个最耗时的部分并行化， 顺序程序，检测片段之间的重叠， 构建重叠群的片段比对。 平行序列 汇编程序将能够使用许多 计算机将成千上万的DNA片段组合成序列 低错误。 研究者将改进多重序列 通过寻址阅读帧移位的对准程序， 蛋白质、cDNA和基因组DNA序列。

DESCRIPTION (Adapted from the Investigator's Abstract): The investigator will develop a software tool for assembling DNA fragments generated in megabase- scale shotgun sequencing projects. The software will be tested first on DNA fragments generated by computers from megabase DNA sequences and then on real DNA fragments from large-scale sequencing projects. The software will be freely distributed to nonprofit organizations. The investigator will assist the integration of the software into sequencing environments at genome centers. The objective of this project will be achieved by making two major improvements to a DNA sequence assembly program developed previously. The first improvement is to develop a strategy for solving the problems caused by repetitive sequences. In this strategy, all the fragments from a repetitive sequence are identified, and the uncertainties in assembly of the fragments are resolved using additional information on the fragments that flank copies of the repetitive sequence. The second improvement is to increase the capacity of the assembly program by developing a parallel version of the program in the PVM parallel programming environment on a local network of computers. The investigator will parallelize the two most time-consuming parts of the sequential program, the detection of overlaps among fragments and the construction of fragment alignments for contigs. The parallel sequence assembly program will be able to use the computation power of many computers to assemble tens of thousands of DNA fragments into sequences of low error. The investigator will improve the multiple sequence alignment program by addressing reading frame shifts in comparison of protein, cDNA and genomic DNA sequences.