RESEARCH-PGR: Illuminating the Plant Protein Interactome with Genome-wide Signatures of Coevolution

植物基因组测序已经打开了一个庞大的基因目录，但我们对每个基因的具体作用的理解远远落后，这意味着功能只分配给了已知植物基因的一小部分。单个基因很少单独工作，特定基因功能的一个关键方面是其相互作用的伙伴。破译整个植物基因组的遗传相互作用将促进我们对大量未表征基因功能的理解；然而，大多数现有的识别基因相互作用的方法都是费力的，昂贵的，并且难以扩展到整个基因组。该项目将开发可扩展的、高效的计算分析，以确定基于植物基因组中基因特异性进化速率的基因相互作用。这些计算工具将成为适用于任何植物类群的资源，并用于为几个科学和农业上重要的植物类群生成遗传相互作用数据库，植物遗传学家和分子生物学家可以通过这些数据库来搜索感兴趣的基因的新功能。研究结果还将用于探索整个遗传相互作用网络的涌现特性，这将阐明植物基因组的更高层次功能结构。这些资源的影响将通过提供研讨会，为有兴趣使用这些工具的研究人员提供培训，并通过与一年级本科生接触，扩大在计算生物学领域的参与，从而最大化。遗传相互作用是基因功能的重要指标，但现有的鉴定植物相互作用全基因组组合（即相互作用组）的工具仅提供了部分视图。相互作用蛋白之间的进化速率共变（ERC）的系统发育特征已成功应用于植物中的小组基因，以证明酶复合物中已知亚基之间的共同进化，并在非植物谱系中全基因组的所有基础上发现新的相互作用伙伴。然而，ERC从未在植物全基因组规模上应用，这可能是由于植物中发生的特别频繁的基因和基因组复制引起的系统发育挑战。该项目将为植物基因组的ERC分析开发一个新的管道，该管道将采用现有的理论来协调基因树之间复杂的复制历史。该管道将用于在植物进化的多个层面上进行ERC分析，并将结果编译成基于网络的植物相互作用组数据库，该数据库可用于查询特定的遗传相互作用。此外，由此产生的基于ERC的相互作用数据将与现有的结合分析和基于共表达的相互作用相结合，以功能性地验证ERC结果。最后，网络分析将识别相互作用的协同功能蛋白模块，并确定网络结构是如何由基因功能驱动的。这项工作将创造一种资源，使基因组生物学家能够为基因组注释工作添加所需的信息层，并为分子生物学家提供丰富的准备测试的功能假设。所有项目成果都可以通过CyVerse和长期存储库（如GitHub和Dryad）免费访问。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Sequencing plant genomes has unlocked a massive catalog of genes and yet our ability to understand the specific role of each gene lags far behind, meaning functions have only been assigned to a fraction of known plant genes. Individual genes rarely work alone, and a critical aspect of a given gene's function is its interacting partners. Deciphering genetic interactions throughout plant genomes will advance our understanding of the function of a large number of uncharacterized genes; however, most existing methods for identifying genetic interactions are laborious, expensive, and difficult to scale to the entire genome. This project will develop scalable, efficient computational analyses tailored to identifying genetic interactions based on gene-specific rates of evolution across plant genomes. These computational tools will serve as resources applicable to any group of plants and be used to generate databases of genetic interactions for several scientifically and agriculturally important plant groups, which plant geneticists and molecular biologists can search to identify new functions for genes of interest. The results will also be used to probe the emergent properties of entire networks of genetic interactions, which will illuminate the higher-level functional architecture of plant genomes. The impact of these resources will be maximized by offering workshops that will provide training to researchers interested in using these tools and by engaging with first-year undergraduates to broaden participation in the field of computational biology.Genetic interactions are an important indicator of gene function but the existing tools for identifying the genome-wide assemblage of interactions (i.e., the interactome) in plants provide only a partial view. The phylogenetic signature of evolutionary rate covariation (ERC) between interacting proteins has been successfully applied on small sets of genes in plants to demonstrate coevolution between known subunits within enzyme complexes and on a genome-wide all-by-all basis in non-plant lineages to discover novel interaction partners. However, ERC has never been applied at genome-wide scale in plants, likely due to phylogenetic challenges caused by the especially frequent gene and genome duplication that occur in plants. This project will develop a novel pipeline tailored to ERC analyses of plant genomes, which will employ existing theory on reconciling complex histories of duplication across gene trees. This pipeline will be used to perform ERC analyses at multiple levels of plant evolution and compile the results to generate a web-based plant interactome database, which can be queried for specific genetic interactions. Further, the resulting ERC-based interaction data will be integrated with existing binding assay and coexpression-based interactions to functionally validate ERC results. Finally, network analytics will identify modules of interacting cofunctional proteins and determine how network structure is driven by gene function. This work will create a resource that will empower genome biologists to add a needed layer of information to genome annotation efforts and provide molecular biologists with a wealth of ready-to-test functional hypotheses. All project outcomes will be freely accessible through CyVerse and long-term repositories such as GitHub and Dryad.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.