Embryonic gene regulatory networks from spatially resolved transcriptomes

 描述（由申请人提供）：系统发育生物学的一个主要目标是创建描述、模拟和预测胚胎发生期间基因调控相互作用的完整补充的数据集和模型。这些数据集和模型对于充分理解基因组信息如何转化为解剖结构至关重要。还原论的方法已经确定了几十个基因之间的相互作用，但技术上的限制阻碍了在高空间分辨率和全基因组尺度上阐明整个系统的调控关系。该项目将解决这一挑战，并使用新技术，使大规模诱变和单细胞的全基因组表达谱生成基因调控网络模型。斑马鱼囊胚将被用作脊椎动物模型系统，因为它与哺乳动物胚胎的相似性以及强大的遗传学，成像和基因组方法的适用性。该项目建立在长期合作的基础上，将希耶实验室在发育生物学、成像和遗传学方面的专业知识与Regev实验室在计算生物学、基因组学和系统生物学方面的专业知识相结合。优化的一代CRISPR/Cas9基因组编辑将用于产生在早期胚胎发生期间表达的数十种转录调节因子的突变体。突变体的特征将通过使用称为Seurat的新技术生成全基因组高分辨率基因表达图谱来表征。Seurat将单细胞RNA测序与细胞到胚胎中特定区域和细胞类型的计算映射相结合。所得到的转录组图谱作为输入，使用基于聚类和贝叶斯建模方法生成基因调控网络活性的模型。将通过分析转录调节因子扰动后的基因表达，在体内测试计算机模拟预测的调控相互作用。该项目实现了PAR-15-020的既定目的，“补充现代发育生物学的还原论重点，并提供对导致正常和异常胚胎发生的因果关系的更全面的理解”。该项目中发现的基因调控相互作用将有助于为编程和重编程方法提供信息，并将确定可能参与出生缺陷发展的候选相互作用。该项目将为发育和系统生物学家生成广泛的高质量数据集和地图集，并为剖析其他系统中的基因调控网络提供框架。

 DESCRIPTION (provided by applicant): A major goal of systems developmental biology is to create datasets and models that describe, simulate and predict the full complement of gene regulatory interactions during embryogenesis. Such datasets and models are essential to fully understand how genomic information is translated into anatomical structure. Reductionist approaches have identified interactions among dozens of genes, but technical limitations have hindered the systems-wide elucidation of regulatory relationships at both high spatial resolution and whole-genome scale. This project will address this challenge and use novel technologies that enable large-scale mutagenesis and genome-wide expression profiling of single cells to generate gene regulatory network models. The zebrafish blastula will be used as a vertebrate model system, because of its similarities to mammalian embryos and the applicability of powerful genetic, imaging and genomic approaches. The project builds on a long-standing collaboration that combines the Schier lab's expertise in developmental biology, imaging and genetics with the Regev's lab expertise in computational biology, genomics and systems biology. Optimized one-generation CRISPR/Cas9 genome editing will be used to generate mutants for dozens of transcription regulators expressed during early embryogenesis. Mutants will be characterized by generating whole-genome high-resolution gene expression atlases using a novel technology called Seurat. Seurat combines single-cell RNA sequencing with computational mapping of cells to specific regions and cell types in the embryo. The resulting transcriptome maps serve as the inputs to generate models for gene regulatory network activity using clustering-based and Bayesian modeling approaches. Regulatory interactions predicted in silico will be tested in vivo by analyzing gene expression upon perturbation of transcription regulators. The project fulfills the stated purpose of PAR-15-020 "to complement the reductionist focus of modern developmental biology and provide a more comprehensive understanding of the causal relationships leading to normal and abnormal embryogenesis". The gene regulatory interactions discovered in this project will help inform programming and reprogramming approaches and will identify candidate interactions that might be involved in the development of birth defects. The project will generate extensive high-quality datasets and atlases for developmental and systems biologists and provide a framework to dissect gene regulatory networks in other systems.