Identifying the building blocks of protein structures

超二级结构单元(SSU)被定义为在给定的褶皱中相互堆积的三个或更多二级结构。当这种SSU出现在不同的褶皱中时，它们被称为“乐高”。已经发现了许多乐高积木的轶事例子，如4-a-螺旋束和3-b-角，但还没有系统地尝试探索整个结构数据库，以获得已知蛋白质折叠的重复结构模式的完整列表。这个项目将开发必要的方法来进行这样的搜索，并充分描述由此产生的乐高积木的特征。这样的努力将为蛋白质折叠提供一个全新的视角。具体地说，该项目将试图回答以下问题：1.乐高积木有多大？2.乐高积木可以覆盖多大程度的已知折叠？3.控制乐高与乐高相互作用的规则是什么？4.人们能否通过遵循这些规则来构建新的蛋白质折叠？5.代表同一乐高的序列有多保守？6.乐高与功能位点相关吗？这个项目将是对当前功能基因组学方法的重要补充。大多数目前的方法通过序列和结构与现有折叠的相似性来推断蛋白质的功能。该项目的结果将为建立不同褶皱之间的联系和将功能推断扩展到已知褶皱的边界之外提供基础。该项目将导致一系列基于乐高的算法的开发，用于预测蛋白质的结构和功能。乐高和他们的各种财产将构成一个公开可用的数据库。通过万维网，研究人员将能够执行两种类型的搜索。可以提交一个蛋白质结构来发现它所包含的乐高。用户还可以根据所有乐高图谱搜索蛋白质序列；匹配的乐高可以提供新序列无法获得的信息。

A Supersecondary Structure Unit (SSU) is defined as three or more secondary structures in a given fold that pack against one another. When such SSUs occur in different folds, they are referred to as "legos". Numerous anecdotal examples of legos, such as 4-a-helix-bundle and 3-b-corner, have been identified, but there has been no systematic attempt to explore the entire structural database for a full list of recurrent structural patterns from which know protein folds might be constructed. This project will develop the necessary methods to carry out such a search and to fully characterize the resulting lego set.Such an effort will provide an entirely new view for protein folds. Specifically, the project will seek to answer the following questions: 1. How large is the lego set? 2. To what extent can the lego set cover known folds? 3. What are the rules governing lego-lego interactions? 4. Can one construct novel protein folds by following the rules? 5. How conserved are sequences representing the same lego?6. Do legos correlate to functional sites?This project will be an important complement to current approaches in functionalgenomics. Most current approaches infer protein function by sequence and structural similarities to existing folds. The results obtained from this project will provide a basis to relate different folds and to extend functional inference beyond the boundary of known folds. The project will lead to the development of an array of lego-based algorithms forpredicting protein structure and function. The legos and their various properties will constitute a publicly available database. Via the World Wide Web, researchers will be able to perform two types of searches. A protein structure can be submitted to discover the legos it encompasses. The user can also search a protein sequence against all lego profiles; the matched legos may provideinformation otherwise unavailable for novel sequences.