Understanding drought tolerance with cell-type specific-data sets and computational approaches

21年前，拟南芥基因组序列的发表是植物生物学上具有里程碑意义的事件。这一序列极大地促进了植物研究。然而，在Araport 11基因组中预测的27,655个基因产物中，根据直接实验数据，只有约42%的基因产物具有氧化石墨烯术语，约36%的基因产物仍然没有任何相关的氧化石墨烯分子功能或生物过程术语[2]。尽管已经产生了大量的基因表达数据，但从特定细胞类型中分离RNA的新方法正在允许以特定细胞的分辨率对生物过程进行前所未有的了解，这对于揭示在“散装组织”样本中无法辨别的反应非常重要。为了充分利用这些植物生物学数据集和方法，我提出了一种双管齐下的生物信息学和基于生物学的方法。第一个目标是开发生物信息学工具，使研究人员（包括我实验室的研究人员）能够利用生成的大量信息。在我实验室的生物分析资源网站bar.utoronto.ca上已经有几十个这样的工具，包括一套ePlants[3,4]，它有一个直观的界面，可以浏览从公里到纳米尺度的数据。BAR每月从世界各地的研究人员那里获得400万的页面浏览量，部分原因是我们在这些量表之间建立了联系，以产生假设。我们感兴趣的是在水平和物种之间产生更多的联系，例如，不同物种的同源物中是否存在类似的非同义变化热点，我们是否可以使用越来越精确的结构预测方法来理解这些热点的功能背景？湿实验室组件将旨在了解这种变化在非生物应激信号的背景下。我提出的研究的第二个目标将利用我的实验室从遭受干旱胁迫的植物的保护细胞中产生的新的细胞类型特异性数据集，我们在任何时间点观察到3,973个差异表达基因[5]。优先T-DNA突变体的热成像表明，大约25%的这些候选基因参与了干旱反应。我们将在分子水平上研究这些候选植物，以了解它们如何适应干旱响应途径，并将在农业重要植物中进行转化实验。这项研究的意义将是相当大的：我的实验室开发的网络基础设施将使基于大型数据集的新见解成为可能，将利用这些公共资助数据集的价值，并将允许更好地了解植物。这对于到2050年养活90亿人口至关重要。我们很高兴不仅能继续开发拟南芥ePlant，还能开发其他用于栽培物种的ePlant。该项目的植物部分将产生可用于农业的新型抗逆性基因。

The publication 21 years ago of the genome sequence of Arabidopsis thaliana was a landmark event in plant biology. The sequence has benefitted plant research immensely [1]. Of the 27,655 predicted gene products in the Araport 11 version of the genome, however, only ~42% have a GO term assigned to them based on direct experimental data, and around 36% still do not have any associated GO Molecular Function or Biological Process terms [2]. Although large amounts of gene expression data have been generated, new methods for the isolation of RNA from specific cell types are allowing unprecedented understanding of biological processes at the resolution of specific cells, which is important to uncover responses not discernible in "bulk tissue" samples. To fully take advantage of these plant biology data sets and methods, I am proposing a two-pronged bioinformatic and biology-based approach. The first objective is to develop bioinformatic tools to enable researchers (including those in my lab) to make use of the vast amount of information being generated. Dozens of such tools are already available at my lab's Bio-Analytic Resource website at bar.utoronto.ca, including a suite of ePlants [3,4], which have an intuitive interface for browsing data from the kilometre to nanometre scale. The BAR receives 4M page views per month from researchers worldwide, in part because we have created connections between these scales for hypothesis generation. We are interested in generating more connections between levels and across species - for instance, are there similar hotspots of non-synonymous changes in homologs from different species and can we use increasingly accurate structure prediction methods to understand the functional context of these hotspots? A wet-lab component will aim to understand this variation in the context of abiotic stress signaling. The second objective of my proposed research will leverage new cell-type-specific data sets my lab has generated from guard cells of plants subjected to drought stress, where we observed 3,973 differentially-expressed genes across any timepoint [5]. Thermal imaging of prioritized T-DNA mutants suggests that roughly 25% of these candidates are involved in drought response. We will investigate several of these candidates at the molecular level to understand how they fit into drought response pathways, and will also undertake translational experiments in agriculturally-important plants. The significance of this research will be considerable: cyberinfrastructure developed by my lab will enable new insights based on large data sets, will leverage the value of these publicly-funded data sets, and will permit a better understanding of plants. This is critical to feeding 9 billion people by 2050. We are excited to continue to develop not only the Arabidopsis ePlant but other ePlants for cultivated species. The in planta part of this project will result in novel stress-tolerance genes that can be deployed agriculturally.