Modeling the processing of signaling cues by transcriptional networks during cell-fate choice

在发育过程中，细胞分化以产生形成功能性生物体所需的细胞类型的多样性。在血液发育过程中，单个造血干细胞可分化成12种不同的细胞类型。造血细胞在不同谱系之间做出选择的决定取决于细胞的内部状态（由谱系指定转录因子网络决定）和细胞因子小分子提供的外部信号。此外，这两个系统，细胞因子信号和转录因子网络，并不是相互独立的。该项目旨在结合计算和实验来了解信号和转录因子网络之间的相互作用如何控制细胞命运的选择。这项研究将促进我们对血液分化过程中细胞命运选择的理解，并且这些见解将与其他组织和生物体相关。在研究过程中开发的计算工具将在发育生物学中具有广泛的适用性。教育活动旨在促进本科和研究生阶段生物学学生的定量和建模技能的发展。这将通过以下方式实现：1)开发基于研究活动的课程模块，并将其提供给教学社区；2)指导本科生进行研究项目。细胞自主基因调控网络和细胞外源细胞因子信号传导常常被视为造血过程中细胞命运规范的竞争和相互排斥的假设。研究人员提出，细胞命运是细胞因子信号和基因调控网络之间相互作用产生的一种紧急特性。他们将通过紧密耦合的实验和数学建模活动来验证这一假设。研究计划利用了一个独特的实验工具；造血细胞可以沿着不同的谱系在一个确定的起点诱导分化，以及时探测细胞因子-转录因子的相互作用。目的1将通过测量高时间分辨率的全基因组表达和计算成对互信息，推断参与巨噬细胞-中性粒细胞细胞命运决定从头开始的基因调控网络。目标2将建立信号效应/转录因子网络的微分方程模型，以研究如何产生紧急动态。这些模型将具有预测性，并允许模拟不同细胞因子和扰动的影响。Aims 1和2在网络水平上研究系统动力学，而Aims 3将确定13个转录因子和细胞因子受体位点的核心组如何在DNA序列水平上及时调节。本文将采用一种新的实验计算方法，用于逆向工程顺式调控模块逻辑，以识别远端增强子和消声器，并确定它们是如何被调控的。构建全面的顺式监管模块记者库，及时对记者活动进行统计。时间分辨活性数据将用于约束基于预测序列的转录热力学模型，以确定调节模块的转录因子和蛋白质-蛋白质相互作用。本研究将使用模型系统的独特工具、时间序列数据和建模来确定基因调控网络如何以环境依赖的方式处理细胞因子信号。模型和报告者库将成为广泛的发育生物学家的资源。

During development, cells differentiate to generate the diversity of cell types required to make a functional organism. In blood development, a single hematopoietic stem cell gives rise to about 12 distinct cell types. A hematopoietic cell's decision to choose between alternative lineages depends on both the internal state of the cell, defined by networks of lineage-specifying transcription factors, as well as external signals provided by small molecules called cytokines. Furthermore, the two systems, cytokine signaling and transcription factor networks, do not function independently of each other. This project aims to use a combination of computation and experiment to understand how the interaction between signaling and transcription factor networks controls cell-fate choice. This study will advance our understanding of cell-fate choice during blood differentiation and the insights will be relevant to other tissues and organisms. The computational tools developed during the course of research will have broad applicability in developmental biology. Educational activities aim to promote the development of the quantitative and modeling skills of biology students at the undergraduate and graduate levels. This will be accomplished by 1) developing course modules based on the research activities and making them available to the teaching community and 2) mentoring undergraduate students in research projects. Cell-autonomous gene regulatory networks and cell-extrinsic cytokine signaling have often times been viewed as competing and mutually-exclusive hypotheses for the specification of cell fate during hematopoiesis. The investigators propose instead that cell fate is an emergent property arising from interactions between cytokine signaling and gene regulatory networks. They will test this hypothesis with tightly coupled experimental and mathematical modeling activities. The research plan leverages a unique experimental tool; hematopoietic cells that can be inducibly differentiated along alternative lineages at a defined starting point to probe cytokine-transcription factor interactions in time. Aim 1 will infer the gene regulatory networks involved in the macrophage-neutrophil cell-fate decision de novo by measuring genome-wide expression at high temporal resolution and computing pair-wise mutual information. Aim 2 will build differential equation models of signaling effector/transcription factor networks to investigate how emergent dynamics are produced. The models will be predictive and allow the simulation of the effects of different cytokines and perturbations. While Aims 1 and 2 investigate the system dynamics at the network level, Aim 3 will determine how a core group of 13 transcription factor and cytokine receptor loci are regulated in time at the level of DNA sequence. Here, a novel experimental-computational approach for reverse engineering cis-regulatory module logic will be utilized to identify distal enhancers and silencers and determine how they are regulated. A comprehensive cis-regulatory module reporter library will be constructed and reporter activities will be measured in time. Time-resolved activity data will be used to constrain predictive sequence-based thermodynamic models of transcription to determine the transcription factors and protein-protein interactions regulating the modules. This study will use unique tools of the model system, time series data, and modeling to determine how gene regulatory networks process cytokine signals in a context-dependent manner. The models and reporter library will be a resource for a wide range of developmental biologists.