Predictive transcription factor modeling to program endodermal cell fates

描述（由申请人提供）：在本提案中，我详细介绍了一项计划，通过计算预测理解转录因子如何决定细胞身份，将细胞编程为胰腺和肠道命运。这个5年计划概述了我将如何过渡到一个独立的终身制研究者桥接干细胞，基因组学和计算生物学领域，以追求我的长期目标，使干细胞衍生的细胞和器官移植治疗糖尿病和消化系统疾病。为了实现这些目标，我将听取理查德·马斯博士和大卫·吉福德博士的建议，他们分别是发育和干细胞生物学、计算生物学和基因组学方面的知名专家。计划中的研究和职业发展活动将在布里格姆妇女医院和哈佛医学院进行，我计划利用那里优良的研究和培训环境。操纵细胞身份以从头产生大量胰腺和肠细胞的能力将改变糖尿病和消化系统疾病的研究和治疗。目前从干细胞中获得治疗相关细胞类型的方法是费力和经验性的，我们相信，对转录因子如何指导细胞命运决定的预测性理解将显着提高我们生产治疗相关胰腺和肠细胞用于疾病建模和移植的能力。我们最近开发了一种基于基因组技术的实验-计算管道，称为DNase-Seq，它改变了我们理解动态转录因子结合和功能的能力，我们利用这种管道揭示了一种结合层次结构，解释了转录因子如何填充调控DNA。在这个项目中，我们将开发和测试预测模型，以解释在所有组织中使用的细胞间信号通路如何在内胚层中特异性地起作用，以促进胰腺和肠细胞命运的采用。为了实现这一目标，我们将回答以下问题：（i）我们能否开发一个完全预测的模型来解释先驱转录因子的结合决定，这些转录因子位于转录因子结合的层次结构之上？(ii)我们能否预测胰腺和肠分化过程中关键细胞间信号通路RA和Wnt下游转录因子的动态结合？(iii)我们能确定预测RA和Wnt依赖性转录因子结合DNA时将被激活的基因的规则吗？通过回答这些问题，我们将有工具重新连接维甲酸和Wnt信号通路，以促进直接采用胚胎干细胞的胰腺和肠道祖细胞命运，为预测性细胞治疗提供关键的第一步。

DESCRIPTION (provided by applicant): In this proposal, I detail a plan to program cells to pancreatic and intestinal fates through computationally predictive understanding of how transcription factors dictate cellular identity. This 5-year plan outlines how I will transition toan independent tenure-track investigator bridging the fields of stem cells, genomics, and computational biology in order to pursue my long-term goal of enabling stem cell-derived cell and organ transplant therapy for diabetes and digestive diseases. To attain these goals, I will be advised by Dr. Richard Maas and Dr. David Gifford, well-established experts in, respectively, developmental and stem cell biology and computational biology and genomics. The planned research and career development activities will be carried out at the Brigham and Women's Hospital and Harvard Medical School, where I plan to take advantage of the excellent research and training environment. The ability to manipulate cell identity to produce large numbers of pancreatic and intestinal cells de novo would transform the study and treatment of diabetes and digestive diseases. Current approaches to derive therapeutically relevant cell types from stem cells are laborious and empirical, and we believe that developing a predictive understanding of how transcription factors guide cell fate decisions will significantly improve our ability to produe therapeutically relevant pancreatic and intestinal cells for disease modeling and transplantation. We have recently developed an experimental- computational pipeline based on a genomic technology called DNase-Seq that transforms our ability to understand dynamic transcription factor binding and function, and we have employed this pipeline to reveal a binding hierarchy that explains how transcription factors populate regulatory DNA. In this project, we will develop and test predictive models to explain how intercellular signaling pathways that are used in all tissues act specifically in endoderm to promote adoption of pancreatic and intestinal cell fates. To accomplish this goal, we will answer the following questions: (i) Can we develop a fully predictive model to explain the binding decisions of pioneer transcription factors, which sit atop the hierarchy of transcription factor binding? (ii) Can we develop a predictive understanding of the dynamic binding of transcription factors downstream of the key intercellular signaling pathways RA and Wnt during pancreatic and intestinal differentiation? (iii) Can we identify rules that predict the genes that will be activated when RA and Wnt-dependent transcription factors bind DNA? By answering these questions, we will have the tools to rewire the retinoic acid and Wnt signaling pathways to promote direct adoption of pancreatic and intestinal progenitor fates from embryonic stem cells, providing a crucial first step to predictive cell therapy.