RNA 3D Motif Search, Atlas, and Prediction from Sequence

描述(由申请人提供)：我们建议通过提供改进的方法将3D RNA结构与序列和其他实验数据相结合，使不断增长的实验三维(3D)RNA结构数据对生物医学研究人员更有用。本项目开发的新注释工具和服务将纳入核酸数据库，该数据库将为传播项目成果提供一个平台。在预期的好处中，有更好的方法用于1)从序列中预测功能RNA基序的3D结构，2)在基因组中搜索非编码RNA基因，以及3)改善同源RNA序列的比对。我们将注意力集中在重复的、模块化的RNA3D基序上，这些基序存在于各种结构的RNA分子中，并使RNA具有独特的3D形状。这包括发夹环、内部环、连接环和第三交互主题。我们将开发系统的方法来识别、分类和命名重复出现的RNA3D基序，并定义搜索标准，以可靠地在3D结构中找到每个基序的实例。将建立一个注释程序，以便在新结构中快速确定新的基序，并与RNA本体联盟的其他成员合作进行审查。所有实验的RNA3D结构都将用基序列表进行注释。将创建一个Motif地图集，以便向用户提供有关结构中3D Motif实例的信息。这一载有基序注释的新地图集将被添加到核酸数据库(NDB)，这是一个包含含有大分子的核酸的结构和功能注释的网络资源。将制定一个更新程序，以便在PDB档案中有新的RNA结构时，将Motif数据和Atlas条目自动添加到NDB。我们将扩展NDB的查询能力，为用户提供工具，使用多标准搜索NDB中的RNA基序，并将搜索结果与实验置信度衡量相结合。我们将维护有关基序和碱基配对相互作用的统计数据，纳入实验置信度衡量标准，并将这些数据作为改进和验证工具的资源。基序图谱中的每个条目将包括基序的所有实例的结构比对，以揭示每个基序的序列变体，包括插入和缺失的模式。这些数据将与Watson-Crick和非Watson-Crick碱基对的统计协变数据以及碱基堆积和碱基-骨架相互作用的统计数据相结合，以开发每个模块RNA 3D基序的序列可变性的概率模型。这些模型将用于部署一个基于网络的工具，供用户从Motif Atlas中找到与他们提交的发夹、内部或连接环路序列最匹配的3D Motif。 与公共健康相关：最近的研究表明，人类基因组的大部分是转录的，产生的大部分RNA是非蛋白质编码的，其中很大一部分对人类的生殖、生长和发育至关重要。这项建议旨在通过提供改进的方法将3D RNA结构与序列和其他实验数据相结合，使不断增长的实验三维(3D)RNA结构数据对生物医学研究社区更有用。

DESCRIPTION (provided by applicant): We propose to make the growing body of experimental three-dimensional (3D) RNA structure data more useful to biomedical researchers by providing improved methods to integrate 3D RNA structure with sequence and other experimental data. New annotation tools and services developed in this project will be integrated into the Nucleic Acid Database (NDB) which will provide a platform for disseminating project results. Among the expected benefits are better methods for 1) predicting 3D structures of functional RNA motifs from sequence, 2) searching for non-coding RNA genes in genomes, and 3) improving alignments of homologous RNA sequences. We focus attention on recurrent, modular RNA 3D motifs, which occur in a wide variety of structured RNA molecules, and which give RNA its distinctive 3D shape. This includes hairpin loops, internal loops, junction loops, and tertiary interaction motifs. We will develop systematic methods to identify, classify, and name recurrent RNA 3D motifs and to define search criteria to reliably find instances of each motif in 3D structures. An annotation procedure will be established so that new motifs are rapidly identified in new structures and vetted in collaboration with other members of the RNA Ontology Consortium. All experimental RNA 3D structures will be annotated with lists of motifs. A Motif Atlas will be created to make information about 3D motif instances in structures available to users. This new Atlas containing the annotation of motifs will be added to the Nucleic Acid Database (NDB), a web resource containing structural and functional annotation of nucleic acid containing macromolecules. An update procedure will be developed such that motif data and Atlas entries will automatically be added to the NDB as new RNA structures become available in the PDB archive. We will extend the query capabilities of the NDB with tools for users to search the NDB for RNA motifs using multiple criteria and to integrate search results with experimental confidence measures. We will maintain statistics on the occurrences of motifs and base pairing interactions, incorporating experimental confidence measures, and make these data available as a resource for refinement and validation tools. Each entry in the Motif Atlas will include a structural alignment of all instances of the motif to reveal sequence variants for each motif, including patterns of insertions and deletions. These data will be combined with statistical covariation data for Watson-Crick and non-Watson-Crick basepairs and statistical data for base-stacking and base- backbone interactions to develop probabilistic models for the sequence variability of each modular RNA 3D motif. These models will be used to deploy a web-based tool for users to find the 3D motif from the Motif Atlas which best matches the sequences of hairpin, internal, or junction loops that they submit. PUBLIC HEALTH RELEVANCE: Recent work shows that most of the human genome is transcribed, most of the produced RNA is non-protein coding, and a large fraction of it is critical for human reproduction, growth, and development. This proposal aims to make the growing body of experimental three-dimensional (3D) RNA structure data more useful to the biomedical research community by providing improved methods to integrate 3D RNA structure with sequence and other experimental data.