A Next Generation Biological Database Management System for Search Intensive Discovery in Genomics and Proteomics

数据库服务是基因组学中一个爆炸性的问题，并且已经成为蛋白质组学中的一个关键瓶颈。为了解释蛋白质组学实验--使用质谱仪对数百或数千种蛋白质进行测量--研究人员组装了一大批计算机来处理单个质谱仪产生的数据，这并不少见。这项提议资助的研究将产生一个可扩展的数据库管理系统，专门为支持对生物数据类型的复杂查询而设计，包括蛋白质组学的核心序列和质谱学数据。一个中心主题是(1)基于树的存储结构，以使通过数据库检索识别蛋白质所需的相似性搜索更快，以及(2)开发结构化查询语言(SQL)的扩展，以使蛋白质识别算法能够更好地使用。这应该会导致更快的生物信息学算法，使复杂的生物信息学数据库能够快速开发，就像SQL使非常大的商业系统能够开发有效的程序一样。

Database services are an exploding concern in genomics, and are already a critical bottleneck in proteomics. In order to interpret proteomics experiments, in which measurements on hundreds or thousands of proteins are made using mass spectrometers, it is not unusual for researchers to assemble a large farm of computers to process the data produced by a single mass-spectrometer. The research funded by this proposal will result in an extendable database management system designed specifically to support complex inqueries on biological data types, including the sequence and mass spectra data central to proteomics. A central theme is (1) an focus tree-based storage structures to make the similarity searches necessary to identify proteins through database retrieval much faster and (2) the development of extensions to Structured Query Language (SQL) to enable better use of protein identification algorithms. This should lead to much faster bioinformatics algorithms, enabling rapid development of complex bioinformatics databases just as SQL has enabled development of effective programs for very large business systems.