Searching Genomes for Non-Coding RNAs by Their Structure

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): RNA molecules and their functions are central to cellular and molecular biology; their functions range from the classic messengers of the central dogma, to ribozymes that carry out enzymatic activities, to acting as regulatory molecules. The existence and important of non-coding RNAs poses a challenge to bioinformatics. How are biologists to identify these molecules in newly sequenced genomes or sequence databases? 1 key approach is to use bioinformatic methods that can search genomes and databases for specific RNA secondary structures. The research we propose is directed towards developing computational tools to model the structures of RNAs, to perform RNA structure alignment and database searches, and then to experimentally test our predictions. Our approach is based upon conformational graph models and graph theoretic techniques, developing fast, yet accurate, RNA structural homology search tools based upon the notion of graph tree decomposition. These methods can describe both stem-loop and pseudoknot structures. Our preliminary results include successful searches of prokaryotic and eukaryotic genomes for large and complex RNAs by their structure. The specific steps of the proposed work are: (1) development of the conformational graph-tree decomposition method into high throughput tools for RNA structure search that biologists can readily use; (2) development of tools for automated comparative analysis and modeling of non-annotated, unaligned, RNA sequences; (3) application of our method to discovery and annotation of specific RNA gene families, including experimental verification of the predictions; (4) distribution of the search and modeling tools, including a user-friendly interface, together with a conformational graph structure profile database. Our goal is to develop a practical RNA structure modeling and search tool set that biologists can readily use to find RNA structures in genomes or databases, to help them generate testable hypotheses about the numbers, functions and evolution of RNA gene families. RNA molecules are at the center of basic biology as well as public health, and may be a key component of future medical practice. Our proposed research will help biologists find RNA molecules of interest in the mass of genome sequence data being generated.