Computational Proxies for Genomic Databases

[224726]库欣SDB/RUI：基因组数据库的计算代理这是一项为期三年的持续资助的第一年。在理想情况下，一个程序的输出可以很容易地用作另一个程序的输入。然而，科学计算的世界远非完美，科学家们被编写特殊用途的数据转换程序的任务所困扰。业务数据处理通过使用共享数据库来解决互操作性问题。不幸的是，当前的数据库系统不能很好地服务于科学应用程序。本研究通过将数据库连接到计算生物学程序来解决科学领域的互操作性问题。它建立在以前由俄勒冈州波特兰市的俄勒冈研究生院联合开展的面向对象计算化学数据库研究的基础上；华盛顿州里奇兰的巴特尔太平洋西北实验室；纽约冷泉港实验室。通过使用T4噬菌体数据（可能是任何个体生物遗传信息的最接近完整的表示），先前的研究结果正在扩展到涵盖基因组数据库。基因组数据库被扩展为包含一个“计算代理机制”，它允许T4数据的参数化接口到现有的计算密集型分布式应用程序。该研究有三个目标：(1)证明以前的数据库工作可推广到计算化学以外的科学领域，并可作为与所谓的“遗留”应用程序的数据库接口模型；(2)为T4研究人员提供计算操作和基因组数据；(3)证明以前在酵母中进行的基因组数据库工作可转移到其他位点和生物体。这项研究的长期影响是将计算科学家认为过于复杂的当前计算环境转变为以数据为中心的方法，从而提高科学数据和程序的互操作性。＊＊＊

9224726 Cushing SDB/RUI: Computational Proxies for Genomic Databases This is the first year funding of a three-year continuing award. In a perfect world output from one program could easily be used as input to another. The world of scientific computing is far from perfect, however, and scientists are plagued with the task of writing special purpose data conversion programs. Business data processing are solving interoperability problems by using shared databases. Unfortunately, current database systems do not serve scientific applications well. This research addresses the problems of interoperability in the scientific realm by interfacing databases to computational biology programs. It builds on previous research in object-oriented databases for computational chemistry conducted jointly by the Oregon Graduate Institute in Portland, Oregon; Battelle Pacific Northwest Laboratories in Richland, Washington; Cold Spring Harbor Laboratories, New York. Results of previous research are being extended to cover genomic databases by using T4 bacteriophage data (perhaps the most nearly complete representation of genetic information for any individual organism). The genomic database is extended to include a "computational proxy mechanism", which allows for parameterized interface of the T4 data to existing computation-intensive distributed applications. The research has three objectives: (1) to demonstrate that previous database work is generalizable to scientific domains other than computational chemistry, and serviceable as a model for database interface with so-called "legacy" applications; (2) to make computational operations as well as genomic data available to the T4 researchers; and (3) to demonstrate that previous genomic database work in yeasts is transferable to other sites and organisms. The long term impact of this research is to transfer the current computing environment that the computational scientists find too complex to a data-centered approach that improves the interoperability of scientific data and programs. ***