Collaborative Research: Ideas Lab: Discovery of Novel Functional RNA Classes by Computational Integration of Massively-Parallel RBP Binding and Structure Data

基因组测序的进展表明，哺乳动物基因组的很大一部分是从DNA转录到RNA，但不翻译成蛋白质；这些被称为非编码RNA(NcRNAs)。许多经典的ncRNAs在生物学中发挥着重要作用，包括翻译(tRNAs、rRNAs)、剪接(SnRNAs)、转录后基因调控(MiRNAs)等许多生物学过程。虽然这些已知的ncRNA对所有生命都很重要，但它们可能只是ncRNA冰山一角。事实上，我们预计有许多ncRNA类仍然没有特征化。这些物质被称为基因组的“暗物质”，因为我们不知道它们可能在生物学上扮演什么角色。与此同时，蛋白质研究已经确定，数以千计的人类蛋白质与RNA结合。然而，目前尚不清楚这些RNA结合蛋白(RBPs)中有多少与ncRNAs相互作用，以及它们可能与哪些特定的ncRNAs相互作用。我们的目标是通过超大规模的RNA-蛋白质结合分析结合计算分析来解决这两个问题，以整体发现新的ncRNA类别。我们将识别与限制性商业惯例强烈相互作用的特定RNA组，开发定义相互作用特异性的模型，并建立分类系统，根据序列和结构数据预测相互作用。我们还将为我们的发现创建一个网络可访问的数据库，允许任何人访问这些数据，并为本科生提供培训，目标是增加科学领域的性别多样性。我们期望揭示新的ncRNA类的生物学功能，这将为生物技术的发展奠定基础。在过去的十年里，以RNA为中心的蛋白质组学方法，如交联和免疫沉淀(CLIP)以及相关方法，使得对RNA-蛋白质相互作用的前所未有的探索成为可能。这些努力极大地增加了已识别的限制性商业惯例的数量，目前有4,000种人类蛋白质(约占人类蛋白质组的20%)被UniProt注释为“RNA结合”。然而，由于CLIP方法一次只能定位一个蛋白质，所以探索数千个注释的限制性商业惯例是具有挑战性的。因此，像ENCODE这样的联盟工作既耗时又昂贵，而且仅限于绘制人类蛋白质组中的一小部分限制性商业惯例。因此，用目前的方法几乎不可能建立一个全面的RBP-ncRNA相互作用体。我们将使用一种新开发的、高度多元化的方法，在一个实验中生成数百个限制性商业惯例的转录组范围的测量结果。我们将把它与尖端的计算和进化策略结合起来，共同发现和分类新的ncRNA类别。我们的目标是在人类转录组中全面发现和表征新的ncRNAs类别，并以现有方法无法实现的方式评估它们的系统发育。为了实现这一目标，我们将开发新的实验方法和综合计算管道，通过结合已知和新的RNA-蛋白质相互作用来系统地识别新类别的ncRNAs，并发现多价相互作用的簇。我们将识别保守的序列和结构基序，以及特定于小说类的进化模式，并开发计算系统从这些数据中识别小说类的成员。该奖项是由NSF生物科学理事会的四个部门共同赞助的IDEAS实验室的结果。该奖项将由分子和细胞生物科学部、环境生物学部和新兴前沿计划共同资助。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Advances in genome sequencing have revealed that large parts of mammalian genomes are transcribed from DNA to RNA but are not translated into protein; these are referred to as non-coding RNAs (ncRNAs). Many classical ncRNAs play fundamental roles in biology including translation (tRNAs, rRNAs), splicing (snRNAs), post-transcriptional gene regulation (miRNAs), and many other biological processes. While these known ncRNAs are important for all life, they may be just the tip of the ncRNA iceberg. In fact, we expect that there are many ncRNA classes that remain uncharacterized. These are referred to as the ‘dark matter’ of the genome because we don’t know what biological roles they may play. In parallel, protein studies have determined that thousands of human proteins bind to RNA. Yet it remains unknown how many of these RNA binding proteins (RBPs) interact with ncRNAs, and which specific ncRNAs they might interact with. Our goal is to tackle both problems using very large-scale RNA-protein binding assays combined with computational analysis to uncover new classes of ncRNAs en masse. We will identify specific groups of RNAs that interact strongly with RBPs, develop models that define interaction specificity, and classification systems to predict interactions from sequence and structural data. We will also create a web-accessible database of our findings, allowing anyone to access the data, and train undergraduates with the goal of increasing gender diversity in science. We expect to reveal the biological functions of novel ncRNA classes, which will lay the foundation for biotechnology development.Over the past decade, global RNA-centric proteomics methods like crosslinking and immunoprecipitation (CLIP) and related approaches have enabled unprecedented exploration of RNA-protein interactions. These efforts have vastly expanded the number of identified RBPs, with 4,000 human proteins (~20% of the human proteome) currently annotated as “RNA-binding” by UniProt. However, because CLIP approaches can only map a single protein at a time, it is challenging to explore the thousands of annotated RBPs. As a result, consortium efforts like ENCODE are time-consuming and expensive, and have been limited to mapping a fraction of the RBPs in the human proteome. Thus, the creation of a comprehensive RBP-ncRNA interactome is near impossible with current approaches. We will use a newly developed, highly multiplexed approach to generate transcriptome-wide measurements across hundreds of RBPs in a single experiment. We will combine this with cutting edge computational and evolutionary strategies to uncover and classify novel classes of ncRNAs en masse. Our goal is to comprehensively discover and characterize novel classes of ncRNAs in the human transcriptome and assess their phylogeny in a way that is impossible using existing methods. To achieve this goal, we will develop novel experimental methods and integrative computational pipelines that will systematically identify novel classes of ncRNAs by combining both known and novel RNA-protein interactions and uncover clusters of multivalent interactions. We will identify conserved sequence and structural motifs, and evolutionary patterns specific to the novel classes, and develop computational systems to recognize members of the novel classes from these data.This award was the result of an Ideas Lab that was co-sponsored by the four divisions in the NSF Directorate of Biological Sciences. It will be co-funded by the Division of Molecular and Cellular Biosciences, the Division of Environmental Biology, and the Emerging Frontiers program.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.