SCOP: A Revitalized Foundation for Studies of Protein Structure and Evolution

项目概述/摘要该项目的主要目标是恢复蛋白质结构分类（SCOP）和ASTRAL数据库，以便更好地服务于当前用户和更大的科学界的需求。这两个数据库都提供了精心策划的资源，被生物学家广泛用于探索感兴趣的蛋白质的远程同源物，并被计算生物学家作为基准预测算法的“黄金标准”。然而，自早期版本以来，这两个数据库都没有改变其基本设计——SCOP是15年前的版本。我们将重新设计这两个数据库的内部结构，以解释在数据库最初创建时未被认识到的蛋白质进化方面，例如变形蛋白和进化出不同折叠的同源蛋白。我们还将为这两个数据库开发一个统一的接口，使科学家能够轻松地找到并关注感兴趣的蛋白质或家族，因为随着结构的增加，当前的分层视图越来越笨拙。由于SCOP管理的过程由于大量的结构被解决而成为瓶颈，我们计划构建自动化工具来协助分类。我们还将创建接口，允许生物学家使用后一种工具提交序列或结构进行自动分类。在许多情况下，这将使结构生物学家能够在新解决的结构发表之前深入了解蛋白质的进化或功能。

DESCRIPTION (provided by applicant): Project Summary/Abstract The primary goal of this project will be to revitalize the Structural Classification of Proteins (SCOP) and ASTRAL databases, both in order to better serve the needs of both current users and the larger scientific community. Both databases provide carefully curated resources that are widely used by biologists to explore remote homologs of proteins of interest, and by computational biologists as a "gold standard" for benchmarking prediction algorithms. However, neither database has changed its basic design since early releases-15 years ago in the case of SCOP. We will redesign the internals of both databases in order to account for aspects of protein evolution that were not appreciated at the time the databases were first created, such as metamorphic proteins and homologous proteins that have evolved different folds. We will also develop a unified interface to both databases that will allow scientists to easily find and focus on proteins or families of interest, as the current hierarchical view is increasingly unwieldy as more structures are added. Since the process of SCOP curation has become a bottleneck due to the large number of structures being solved today, we plan to build automated tools to assist in the classification. We will also create interfaces to allow biologists to submit sequences or structures for automated classification using the latter tools. In many cases, this would enable structural biologists to gain insight into a protein's evolution or function prior to publication of a newly solved structure. PUBLIC HEALTH RELEVANCE: PROJECT NARRATIVE The SCOP and ASTRAL databases are carefully curated resources that are widely used by biologists to explore the structure, function, and evolution of protein families of interest, and by computational biologists as a "gold standard" for benchmarking prediction algorithms. Protein structures are essential to many modern studies of pathogens and human diseases, and medical conditions are being rapidly linked to specific mutations. Although the flood of structural information threatens to overwhelm our current capacity for analysis, our proposed changes to the curation procedures for SCOP and ASTRAL, and improvements to the underlying structure of the databases, will allow these resources to continue to yield biological and medical insight for many years to come.