DATABASES FOR MOLECULAR MODELING

我们已经开发了用于分子建模的数据库和软件 以及序列-结构关系的分析。这些工具是 免费分发给生物学家和生物技术开发人员 软件。工作集中在三个方面：1)持续发展 面向对象的研究数据库，2)MMDB的开发， 用于建模应用程序的三维结构数据库，以及3) 开发MMDB-API，一个应用程序编程接口。PKB有 已修改为包括显式系综模型和相关无序模型 从蛋白质数据库(PDB)文件中解析，经过验证检查， 以便将该信息合并到MMDB中。稳定的结构 根据稳定研究的需要，现在也正在分配标识符 数据集标识和MMDB增量更新。MMDB是ASN.1 数据库，其规范精确地定义了 生物聚合物和非聚合物组分，并明确映射到3- 维度坐标。已经有几个新的验证检查 整合到MMDB中，包括异常绑定的标志和 “标准”基团的手性，反常原子符号的修正 在频繁使用的情况下使用，并遗漏明显错误的PDB焊接数据。 对NCBI序列数据库和Medline的交叉引用也被 添加到MMDB中，并且现在产生了用于“Entrez”的PDB派生序列 直接从它那里。MMDB数据库本身现在作为 在网络和万维网Entrez中可搜索的“结构化”数据库。 MMDB-API是从MMDB自动填充的内存中数据结构 文件和相关的C函数库。它已用于 为Entrez生成Kinemage和PDB报告生成器，以及测试版 Entrez图形查看器的。这项工作的意义在于 为生物学家提供轻松访问结构数据和信息的途径 关于已知的序列-结构关系，并在提供一种软件 为分子建模软件的开发人员提供基础设施。

We have developed databases and software useful for molecular modeling and analysis of sequence-structure relationships. These tools are distributed freely to biologists and developers of biotechnology software. Work has concentrated in three areas: 1) continued development of the "PKB" object-oriented research database, 2) development of MMDB, a 3-dimensional structure database for modeling applications, and 3) development of MMDB-API, an applications programming interface. PKB has been modified to include explicit ensemble and correlated disorder models parsed from Protein Data Bank (PDB) files, subject to validation checks, so that this information may be incorporated into MMDB. Stable structure identifiers are also now being assigned, as required for stable research data set identification and MMDB incremental update. MMDB is an ASN.1 database whose specification defines precisely the covalent structure of biopolymers and non-polymer components and an unambiguous mapping to 3- dimensional coordinates. Several new validation checks have been incorporated into MMDB, including flags for anomalous bonding and chirality in "standard" groups, correction of anomalous atomic symbol usage in frequent cases, and omission clearly erroneous PDB bonding data. Cross references to NCBI sequence databases and Medline have also been added to MMDB, and PDB-derived sequences for "Entrez" are now produced directly from it. The MMDB database itself is now distributed as the searchable "Structure" database in network and world-wide-web Entrez. MMDB-API is an in-memory data structure populated automatically from MMDB files and an associated function library in C. It has been used to produce Kinemage and PDB report generators for Entrez, and a beta version of an Entrez graphics viewer. The significance of this work is in providing biologists with easy access to structural data and information on known sequence-structure relationships, and in providing a software infrastructure to developers of molecular modeling software.