Development and assessment of methods for membrane protein structure prediction

在进化过程中，蛋白质通常保留共有的三维结构特征，即使它们组成的氨基酸序列可能发生显着差异。这些关系也可以在单个蛋白质结构内部表现出来，起源于定义元素的复制和重复。因此，鉴定两种蛋白质之间或同一蛋白质内的两个不同区域之间的进化关系，尽管存在很大的进化距离，仍然具有巨大的价值。随着实验确定的结构的可用性，这种关系可以通过叠加公共区域来揭示，这种技术被称为结构对齐。虽然这些结构对齐方法已被用于探索个体结构内的关系并揭示对称性或伪对称性，但它们具有局限性。值得注意的是，它们不能解释蛋白质在细胞膜内的定位或其构象变化。 为了应对这些挑战，我们已经建立了检测对称性的方法，并对其进行了调整，以提高其对膜蛋白的适用性。这项工作的初步结果被编译成一个名为EncoMPASS（结构和对称性分析膜蛋白百科全书）的数据库。我们继续花费相当大的努力来改进生成EncoMPASS数据库的程序，扩大可以分析的结构库，同时引入更高水平的自动化和计算效率。在NINDS生物信息学核心（Yavaktar和Kumar）的帮助下，我们专注于改善目前通过https://encompass.ninds.nih.gov托管的公共网络服务器提供的数据的可视化。我们的首要目标是创建一个用户友好的资源，可以随时更新最新的数据，并在未来很容易地将新功能纳入其中。

During the course of evolution, proteins often retain shared three-dimensional structural features, even as the sequence of amino acids that they consist of can undergo significant divergence. These relationships can also manifest internally within a single protein structure, originating from duplications and repetitions of defined elements. Consequently, the identification of evolutionary relationships between two proteins, or two different regions within the same protein, despite substantial evolutionary distance, holds immense value. With the availability of experimentally-determined structures, such relationships can revealed by superposing common regions, a technique referred to as structure alignment. While these structural alignment methods have been adapted to explore relationships within individual structures and to uncover symmetries or pseudo-symmetries, they come with limitations. Notably, they do not account for a protein's positioning within the cellular membrane or its conformational changes. To address these challenges, we have taken established approaches for detecting symmetry and tailored them to enhance their applicability to membrane proteins. The initial outcomes from this work were compiled into a database called EncoMPASS (Encyclopedia for Membrane Proteins Analyzed by Structure and Symmetry). We have continued to spend considerable efforts improving the procedures for generating the EncoMPASS database, broadening the pool of structures that can be analyzed, while also introducing greater levels of automation and computational efficiency. With the aid of the NINDS Bioinformatics Core (Yavaktar and Kumar), we have focused on improving the visualization of the data currently available through a public webserver hosted at https://encompass.ninds.nih.gov. We are on the verge of publishing the updated dataset. Our overarching goal is to create a user-friendly resource that can be readily updated with the latest data, and to which new features can easily be incorporated in the future.