Program in Cardiovascular and Metabolic Disorders, Duke-NUS Medical School, 169857, Singapore.
Bioinformatics. 2020 May 1;36(10):3273-3275. doi: 10.1093/bioinformatics/btaa099.
Emerging single-cell RNA-sequencing data technologies has made it possible to capture and assess the gene expression of individual cells. Based on the similarity of gene expression profiles, many tools have been developed to generate an in silico ordering of cells in the form of pseudo-time trajectories. However, these tools do not provide a means to find the ordering of critical gene expression changes over pseudo-time. We present GeneSwitches, a tool that takes any single-cell pseudo-time trajectory and determines the precise order of gene expression and functional-event changes over time. GeneSwitches uses a statistical framework based on logistic regression to identify the order in which genes are either switched on or off along pseudo-time. With this information, users can identify the order in which surface markers appear, investigate how functional ontologies are gained or lost over time and compare the ordering of switching genes from two related pseudo-temporal processes.
GeneSwitches is available at https://geneswitches.ddnetbio.com.
Supplementary data are available at Bioinformatics online.
新兴的单细胞 RNA 测序技术使得捕获和评估单个细胞的基因表达成为可能。基于基因表达谱的相似性，已经开发了许多工具来生成细胞的伪时间轨迹形式的计算排序。然而，这些工具并不能提供一种方法来找到伪时间上关键基因表达变化的排序。我们提出了 GeneSwitches，这是一种工具，可以采用任何单细胞伪时间轨迹，并确定基因表达和功能事件随时间变化的精确顺序。GeneSwitches 使用基于逻辑回归的统计框架来确定基因在伪时间上是开启还是关闭的顺序。有了这些信息，用户可以识别表面标记物出现的顺序，研究功能本体论随时间如何获得或失去，并比较来自两个相关伪时间过程的开关基因的排序。
GeneSwitches 可在 https://geneswitches.ddnetbio.com 上获得。
补充数据可在 Bioinformatics 在线获得。
