DataSheet_1_Revealing Dynamic Mechanisms of Cell Fate Decisions From Single-Cell Transcriptomic Data.pdf

Cell fate decisions play a pivotal role in development, but technologies for dissecting them are limited. We developed a multifunction new method, Topographer, to construct a “quantitative” Waddington’s landscape of single-cell transcriptomic data. This method is able to identify complex cell-state transition trajectories and to estimate complex cell-type dynamics characterized by fate and transition probabilities. It also infers both marker gene networks and their dynamic changes as well as dynamic characteristics of transcriptional bursting along the cell-state transition trajectories. Applying this method to single-cell RNA-seq data on the differentiation of primary human myoblasts, we not only identified three known cell types, but also estimated both their fate probabilities and transition probabilities among them. We found that the percent of genes expressed in a bursty manner is significantly higher at (or near) the branch point (~97%) than before or after branch (below 80%), and that both gene-gene and cell-cell correlation degrees are apparently lower near the branch point than away from the branching. Topographer allows revealing of cell fate mechanisms in a coherent way at three scales: cell lineage (macroscopic), gene network (mesoscopic), and gene expression (microscopic).

细胞命运决定在发育进程中发挥关键作用，但解析该过程的技术仍存在局限。我们开发了一种新型多功能方法Topographer，用于构建单细胞转录组数据的定量沃丁顿景观（Waddington’s landscape）。该方法可识别复杂的细胞状态转变轨迹，并估算以命运概率与转变概率为表征的复杂细胞类型动态变化。同时，其还可推断标记基因网络及其动态变化，以及沿细胞状态转变轨迹的转录爆发（transcriptional bursting）动态特征。我们将该方法应用于原代人成肌细胞分化的单细胞RNA测序（single-cell RNA-seq）数据后，不仅成功识别出三种已知细胞类型，还估算了各类细胞的命运概率及其相互间的转变概率。研究发现，处于（或邻近）分支点的基因中，以爆发方式表达的比例（约97%）显著高于分支前后区域（低于80%）；且分支点附近的基因间与细胞间相关程度均明显低于远离分支的区域。Topographer可从三个尺度以连贯统一的方式揭示细胞命运机制：细胞谱系（宏观尺度）、基因网络（介观尺度）以及基因表达（微观尺度）。