Beyond Pseudotime: Enhanced Single-cell Genomics Tools for Understanding the Temporal Dynamics of Development and Disease

摘要 从一个受精卵开始，人类基因组必须调节一系列令人难以置信的细胞分裂， 命运决定产生成人人体及其约30万亿个细胞[1]。基因组还必须协调 在这些终末细胞类型中具有高度多样化的功能，并且在许多情况下允许对细胞周期的动态响应。 各种刺激-从响应刺激的白色血细胞[2]到响应激素的肝细胞 提示[3]。此外，发育过程是异步的，并且对于许多细胞类型来说， 成年我们理解所有这些过程中的因果联系的基础是时间概念。 虽然时间进程研究在基因组学中有很长的历史[4]，但单细胞基因组技术正在提供 在基因组尺度上对细胞分化和反应的时间动态进行了前所未有的观察[5]。 这将对我们的干细胞治疗策略产生广泛的影响， 发展，以及我们对发育生物学的基本理解。然而，这些推论是迄今为止 有限，并依赖于一个称为“伪时间”的概念[5]，这是很难验证，并可以相对于扭曲 真实的时间。为了真正理解基因组如何协调发育、分化和疾病， 新的工具，使我们能够更好地衡量发展轨迹的几个关键特征： 调节级联的持续时间，发育过程中关键基因组事件的持续时间， 可以调节时间表达模式的特定DNA序列。为了解决这些 关注，我们将开发一套新的工具，利用单细胞读数，以更好地了解 时间的基因调控特别是，我们将专注于高度多重检测，以更好地了解 分化途径中调节级联的必要和充分的排序， 伪时间到真实的时间，以及基因组可扩展的测定，以鉴定和验证确切的调节 定义基因表达时间模式的序列。

ABSTRACT From a single fertilized egg, the human genome must regulate an incredible succession of cellular divisions and fate decisions to give rise to the adult human body and its ~30 trillion cells [1]. The genome must also orchestrate highly diverse functions in these terminal cell types and in many instances allow for dynamic responses to a variety of stimuli - from white blood cells responding to stimulation [2] to hepatocytes responding to hormonal cues [3]. Furthermore, developmental processes are asynchronous and continue for many cell types into adulthood. Fundamental to our understanding of the causal links in all of these processes is the concept of time. While time course studies have a long history in genomics [4], single-cell genomic technologies are providing unprecedented views into the temporal dynamics of cellular differentiation and response at a genomic scale [5]. This will have widespread implications for our strategies of stem cell therapy, windows of intervention in disease progression, and our basic understanding of developmental biology. However, these inferences are to-date limited and rely on a concept called ‘pseudotime’ [5], which is difficult to validate and can be warped relative to real time. To truly understand how the genome coordinates development, differentiation, and disease we need new tools that allow us to better measure several key features of developmental trajectories: the ordering of regulatory cascades, the duration of the key genomic events in developmental processes, and the specific DNA sequences that can regulate temporal expression patterns. In order to address these concerns, we will develop a new suite of tools that leverage single-cell readouts to better understand the genomic regulation of time. In particular, we will focus on highly multiplexed assays to better understand the necessary and sufficient ordering of regulatory cascades in differentiation pathways, assays to convert pseudotime to real time, and genome scalable assays to identify and validate the exact regulatory sequences that define temporal patterns of gene expression.