Over the past decade, single-cell genomics technologies have allowed scalable profiling of cell-type-specific features, which has substantially increased our ability to study cellular diversity and transcriptional programs in heterogeneous tissues. Yet our understanding of mechanisms of gene regulation or the rules that govern interactions between cell types is still limited. The advent of new computational pipelines and technologies, such as single-cell epigenomics and spatially resolved transcriptomics, has created opportunities to explore two new axes of biological variation: cell-intrinsic regulation of cell states and expression programs and interactions between cells. Here, we summarize the most promising and robust technologies in these areas, discuss their strengths and limitations and discuss key computational approaches for analysis of these complex datasets. We highlight how data sharing and integration, documentation, visualization and benchmarking of results contribute to transparency, reproducibility, collaboration and democratization in neuroscience, and discuss needs and opportunities for future technology development and analysis.

在过去的十年中，单细胞基因组学技术允许对细胞类型特异性特征进行可扩展的分析，这大大提高了我们研究异质组织中细胞多样性和转录程序的能力。然而，我们对基因调控机制或控制细胞类型之间相互作用的规则的理解仍然有限。新的计算管道和技术的出现，如单细胞表观基因组学和空间分辨转录组学，为探索生物变异的两个新轴创造了机会：细胞状态和表达程序的细胞内在调控以及细胞之间的相互作用。在这里，我们总结了这些领域中最有前途和最强大的技术，讨论了它们的优势和局限性，并讨论了分析这些复杂数据集的关键计算方法。我们重点介绍了数据共享和集成、记录、可视化和结果基准测试如何促进神经科学的透明度、可重复性、协作和民主化，并讨论了未来技术开发和分析的需求和机会。