Computational prediction of RNA 3D structure using low resolution data and a new

描述（由申请人提供）：大自然已经发展出保护生物体免受外部入侵者（如病毒）或导致疾病（如癌症）的内部功能障碍的机制。其中一种安全机制是基因沉默，它是生物体发育和物种形成的核心。特别是，基因沉默被广泛用于研究人类和模式生物的遗传学。越来越多的证据表明，癌症等疾病是多种基因失调的结果。正常的细胞反应应该是关闭这些基因。然而，在癌症等疾病的情况下，由于未知的原因，细胞无法应对这种情况，导致肿瘤的发展和进展。如果我们希望发现为什么细胞在这些情况下不能保护自己，或者如果我们想有效地干预基因沉默以对抗这些疾病，我们需要更好地理解基因沉默机制。基因沉默是基于一类被称为微rna的核糖核酸。本研究的目的之一是研究这些microrna的分子基础，特别是它们与靶基因的信使rna结合的机制。准确预测microrna的靶基因是目前基因沉默最重要的未解决问题之一。该领域的许多研究由于过度依赖核糖核苷酸链的二级结构而受到阻碍，这严重限制了当前计算模型的有用性和预测价值。核糖核苷酸链的功能主要取决于它们的三级结构。这就是为什么对于目前从事RNA和microrna研究的所有实验室来说，能够快速准确地以原子精度生成RNA三级结构的建模程序是至关重要的。在这项研究中，我建议开发和应用RNA计算机建模框架，该框架建立在我过去20年开发的基础上，以获得microRNAs和靶向信使RNA的高分辨率三级结构。这些三级结构将帮助我们了解microrna如何以及在哪里结合它们的靶标，并准备生物分析来证实这些见解。我们的目标是将这些强大的新工具交到实验人员手中，因为快速可靠地生产RNA结构将对生物医学领域理解和实验这些最近发现的基因沉默途径的能力产生巨大的好处。

DESCRIPTION (provided by applicant): Nature has developed mechanisms to protect living organisms against external intruders such as viruses, or internal dysfunction leading to diseases, such as cancer. One such security mechanism is gene silencing, which is at the heart of organisms< development and speciation. In particular, gene silencing is widely used to study human and model organisms' genetics. Increasing evidence suggests that diseases such as cancer are the result of multiple gene deregulations. A normal cellular reaction would be to shutting down these genes. However, in the case of diseases such as cancer, for unknown reasons the cell is unable to cope with the situation, leading to tumor development and progression. We need to understand better gene silencing mechanisms if we wish to discover why the cell cannot defends itself in these situations, or if we want to intervene efficiently in gene silencing to fight these diseases. Gene silencing is based on a class of ribonucleic acids called microRNAs. One of the aims of this research is to study the molecular basis of these microRNAs, and in particular the mechanism by which they bind the messenger RNAs of targeted genes. Predicting accurately the targeted genes of microRNAs is among the most important unresolved questions of gene silencing at this moment. Many in the field are hindered by their over-reliance on the secondary structures of ribonucleotide chains, which severely limits the usefulness and predictive value of the current computational models. Ribonucleotide chains' function is rather due to their tertiary structure. This is why it is crucial for all laboratories currently engaged in research on RNAs and microRNAs that modeling programs capable of rapidly and accurately producing RNA tertiary structures with atomic precision become available. In this research, I propose the development and application of an RNA computer-modeling framework that builds upon the one I developed over the last 20 years to obtain high-resolution tertiary structures of microRNAs and targeted messenger RNAs. These tertiary structures will help us get insights into how and where microRNAs bind their targets, and prepare biological assays to confirm these insights. We aim to put these new powerful tools in the hands of the experimentalists, since rapidly and reliably producing RNA structures will yield immense benefits to the biomedical field's ability to understand and experiment with these recently discovered gene-silencing pathways. PUBLIC HEALTH RELEVANCE: Accumulating evidence shows that RNA molecules are crucial players in the mechanisms that protect us against external intruders such as viruses, or internal dysfunction that leads to diseases. Determining RNA three-dimensional (3D) structures is key to study RNA function, but is costly and laborious using physical methods. In many ways, this situation hinders the RNA biomedical field, and the research here aims at providing to all computational tools to rapidly and reliably producing RNA 3D structures, yielding immense benefits to the researchers' ability to understand and manipulate RNAs including those in our own protection mechanisms.