Department of Pediatrics, Leiden University Medical Center, 2333 ZA Leiden, the Netherlands;
Department of Pediatrics, Leiden University Medical Center, 2333 ZA Leiden, the Netherlands.
J Immunol. 2020 Aug 1;205(3):864-871. doi: 10.4049/jimmunol.1901530. Epub 2020 Jun 26.
The introduction of single-cell platforms inspired the development of high-dimensional single-cell analysis tools to comprehensively characterize the underlying cellular heterogeneity. Flow cytometry data are traditionally analyzed by (subjective) gating of subpopulations on two-dimensional plots. However, the increasing number of parameters measured by conventional and spectral flow cytometry reinforces the need to apply many of the recently developed tools for single-cell analysis on flow cytometry data, as well. However, the myriads of analysis options offered by the continuously released novel packages can be overwhelming to the immunologist with limited computational background. In this article, we explain the main concepts of such analyses and provide a detailed workflow to illustrate their implications and additional prerequisites when applied on flow cytometry data. Moreover, we provide readily applicable R code covering transformation, normalization, dimensionality reduction, clustering, and pseudotime analysis that can serve as a template for future analyses. We demonstrate the merit of our workflow by reanalyzing a public human dataset. Compared with standard gating, the results of our workflow provide new insights in cellular subsets, alternative classifications, and hypothetical trajectories. Taken together, we present a well-documented workflow, which utilizes existing high-dimensional single-cell analysis tools to reveal cellular heterogeneity and intercellular relationships in flow cytometry data.
单细胞平台的引入激发了高维单细胞分析工具的发展，以全面描绘潜在的细胞异质性。流式细胞术数据传统上通过二维图上的（主观）门控对亚群进行分析。然而，传统和光谱流式细胞术所测量的参数数量不断增加，这就需要将最近开发的许多单细胞分析工具应用于流式细胞术数据。然而，不断发布的新型软件包提供的大量分析选项可能会让计算背景有限的免疫学家感到不知所措。在本文中，我们解释了这些分析的主要概念，并提供了详细的工作流程，以说明它们在应用于流式细胞术数据时的影响和额外的前提条件。此外，我们还提供了易于应用的 R 代码，涵盖了转换、归一化、降维、聚类和伪时间分析，可作为未来分析的模板。我们通过重新分析一个公开的人类数据集来证明我们工作流程的优势。与标准门控相比，我们工作流程的结果在细胞亚群、替代分类和假设轨迹方面提供了新的见解。总之，我们提出了一个有详细记录的工作流程，该流程利用现有的高维单细胞分析工具来揭示流式细胞术数据中的细胞异质性和细胞间关系。
