Moving the Ribosome Database Project (RDP) to a DBMS Foundation

核糖体RNA （rRNA）序列是生物学的核心，从基础和方法学的角度。作为细胞翻译过程中的关键因素，它们本身就是详细的结构/功能研究的主题。此外，它们的进化特征使它们在其他领域，特别是微生物学领域，作为工具非常有用。rRNA序列的比较分析(i)提供了rRNA的二级结构；（ii）导致了普遍生命之树的推断——自然微生物分类的基础——以及古细菌的发现；（iii）为研究微生物生态学的新方法提供了基础；（iv）为医学诊断中许多生物特异性探针的开发确定了目标；提供了环境影响和生物修复评价的新方法。这些研究需要大量的rRNA数据，只有在适当的rRNA序列数据库和相关数据的背景下才能有效地使用这些数据。仅仅提供原始序列数据是不够的；许多分析需要将数据组织成排列，需要有系统发育概述，并且需要工具和（对许多用户来说）服务可用于数据处理和分析。由于没有如此有用、如此通用的rRNA序列数据库，因此在美国国家科学基金会的支持下，伊利诺伊大学建立了核糖体数据库项目（RDP）。通过其电子邮件（server@rdp.life.uiuc.edu）、ftp （rdp.life.uiuc.edu）、gopher （rdpgopher.life.uiuc.edu）和万维网（http://rdpwww.life.uiuc.edu/index.html）服务器，RDP以各种用户可选择的格式提供排列整齐和按系统发育顺序排列的rRNA序列，以及用于处理和分析数据的各种软件。此外，还提供自动序列比对、生物特异性（或群体特异性）探针设计检查、检索用户可选择的序列数据子集和系统发育树以及rRNA二级结构图等服务。RDP目前以文件的形式维护其数据，这些文件的格式适合于RDP和所服务的研究人员使用的工具。随着数据量的增加，显然需要一个更加健壮和结构化的系统。在数据库管理系统（DBMS）上适当设计和实施的RDP版本将：(i)确保RDP的顺利发展；（ii）为RDP数据提供统一而高效的接口，以进行数据积累、管理和查询；（iii）促进增加获取RDP数据和服务的新模式。增加数据结构的目的是将RDP的资源与其他可用数据连接起来，从而促进联邦数据库的创建。除了增加RDP（以及将链接到它的数据）的价值之外，建议的工作将探索有效存储和检索对齐序列数据（RDP的最大组成部分）的方法，以及执行用于管理和利用这些数据的各种复杂查询的方法。本建议的具体目标是：(i)完成基于数据库管理系统的RDP版本的详细设计和实施；（ii）提供有效管理数据所需的图形用户界面；（iii）利用增加的数据结构，进一步自动化区域发展计划的数据更新和管理程序；（iv）透过实施数据库管理系统，提供所有现有的用户访问模式；(v)改善和扩展RDP用户可用的访问模式和查询能力；(5)为RDP与其他数据的整合提供必要的“挂钩”，最初是由密歇根州立大学（MSU）微生物生态学中心（CME）提出的微生物数据库联盟。鉴于目前的rRNA序列数据量（大约3500个序列）和估计的未来水平（数万个序列，或更多），RDP在未来的生物学、医学和环境研究中发挥着至关重要的作用。为了填补这些角色，有必要在不干扰当前服务的情况下更新其设计。本提案解决了这一需求。

Ribosomal RNA (rRNA) sequences are central to biology from both the basic and the methodological perspectives. As the key elements in the cellular translation process they are subjects of detailed structure/function studies in their own right. In addition, their evolutionary characteristics make them extremely useful as tools in other areas, microbiology in particular. Comparative analysis of rRNA sequences has (i) provided the secondary structure of the rRNAs; (ii) led to the inference of the universal tree of life--a basis for a natural microbial classification--and the discovery of the Archaea, (iii) provided the basis for new and powerful approaches to microbial ecology; (iv) defined targets for much of the organism-specific probe development in medical diagnosis; and (v) provided new methods for environmental impact and bioremediation assessment. Such studies require large amounts of the rRNA data, and these can be used effectively only in the context of the appropriate database of rRNA sequences and related data. It is not sufficient to provide just the raw sequence data; many analyses require that the data be organized into alignments, that there be a phylogenetic overview, and that tools and (for many users) ser vices be available for data handling and analysis. In that an rRNA sequence database this useful, this versatile, did not exist, the Ribosome Database Project (RDP) has been established at the University of Illinois with support from the NSF. Through its E-mail (server@rdp.life.uiuc.edu), ftp (rdp.life.uiuc.edu), gopher (rdpgopher.life.uiuc.edu) and World Wide Web (http://rdpwww.life.uiuc.edu/index.html) servers, the RDP offers aligned and phylogenetically ordered rRNA sequences in a variety of user-selectable formats, as well as various software for handling and analysis of the data. In addition, services including automated sequence alignment, organism- specific (or group-specific) probe design checking, retrieval of user-selectable subsets o f the sequence data and phylogenetic trees, and rRNA secondary structure diagrams are provided. The RDP currently maintains its data in files with formats that are appropriate for the tools used by the RDP and by the researchers served. As the volume of data increases, the need for a more robust and structured system is clear. A properly designed and implemented version of the RDP on a database management system (DBMS) will: (i) ensure smooth growth of the RDP; (ii) provide a uniform and efficient interface to the RDP data for data accumulation, curation and querying; and (iii) facilitate the addition of new modes of access to the RDP data and services. The increased structuring of the data is being carried out with an eye toward linkage of the RDP's resources with other available data, facilitating the creation of a federated database. In addition to increasing the value of the RDP (and the data that will be linked to it), the proposed work will explore methods for efficient storage and retrieval of aligned sequence data (the largest component of the RDP), and for performing the kinds of complex queries that are used in curating and exploiting these data. Specific objectives of this proposal are: (i) complete a detailed design and implementation of a DBMSbased version of the RDP; (ii) provide the graphical user interfaces (GUIs) necessary for the efficient curation of the data; (iii) use the increased data structuring to further automate the RDP's data updating and management procedures; (iv) provide all existing user access modes through the DBMS implementation (v) improve and expand the access modes and query capabilities available to users of the RDP; and (v provide the necessary "hooks" for integration of the RDP with other data, initially the federation of microbial databases being proposed by the Center for Microbial Ecology (CME) at Michigan State University (MSU). Given the present volume of rRNA sequence data l approximately 3500 sequences) and the estimated future level (tens of thousands of sequences, or more), the RDP has essential roles to play in biological, medical and environmental research in the future. To fill these roles, it is necessary to update its design without interfering with current services. This proposal addresses this need.