Computational analysis and modeling of noncoding ribonucleic acid structure and function

点击翻译按钮获取中文摘要

Background**Ribonucleic acid (RNA) has diverse cellular functions beyond being a messenger between genes and proteins. Mammalian genomes produce 10 to 20 times more non-coding than protein-coding transcripts. Besides, thousands of human genomic regions that cannot be aligned in primary sequence contain RNA structure. Whether all of these regions and non-coding transcripts are functional is still in question. However, it is now clear that many noncoding RNAs (ncRNAs) are functional and act on the regulation of gene activity. Many ncRNAs have been shown crucial in cell differentiation, reprogramming of human induced pluripotent stem cells, and cell fates during development.**Hypothesis and aims**In particular, large ncRNAs (lncRNAs) create an interaction network that defines the accessibility and flow of genetic information in cells, an address code through which protein complexes, genes, and chromosomes are efficiently placed and retrieved, and carried in appropriate locations. The long-term goal of this research is to decipher this cellular address code, i.e. determine the nature and rules of the organizational system that control molecular accessibility and traffic in cells. If ncRNAs are central to the address code, as accumulating evidences suggest, then establishing the RNA interaction network, and in particular with proteins, is among the first problems to be solved. Biological functions of RNA rely on the inherent nature of its dynamic structural changes that occur at different timescales, which currently often escape biophysical methods such as nuclear magnetic resonance (NMR) and fluorescence spectroscopy. Here, we propose a computational approach that is free of the limitations of experimental techniques to generalize a model of RNA dynamics that is firmly supported by NMR data. Using this model, the specific aims are to:**Aim 1: Determine ncRNA structure and dynamic profile*Aim 2: Determine ncRNA::protein interactions**Approach**Aim 1. We will probe in vivo ncRNA structure using SHAPE (Selective 2'-Hydroxyl acylation Analyzed by Primer Extension) chemistry and cisplatin-RNA cross-linking. SHAPE chemistry identifies flexible RNA regions and cisplatin cross-linking identifies structured RNA regions. We will then determine ncRNA structure and dynamic profile using the computational tools developed in our laboratory (MC-Tools) and the SHAPE and cisplatin data.**Aim 2. We will determine the likelihood of ncRNAs to interact with proteins using a combination of symbolic and numerical computation. We will predict an interaction between a ncRNA and a protein if the ncRNA contains a structural motif previously observed in direct contact with a structural domain of the protein. The direct RNA::protein contacts will be extracted from the experimentally resolved structures of the Protein DataBank. We will then assess the realism of the interaction using molecular docking and chemical context profile matching. We will determine ncRNA interactions from these predictions.**Relevance**This project will improve our understanding of the role of ncRNAs in the regulation of gene expression. It will deliver: (i) structural and dynamic profiles of ncRNAs; and, (ii) new predictions of ncRNA interactions with genomic DNA, other cellular RNAs, and proteins. Collectively, the results of this research will benefit multiple areas of research by providing new insights on gene activity and cellular programs. The combined computational and experimental approach we propose will bring and train together informatics, molecular biology and biochemistry students and personnel.