Xu Bingxian, Ma Dingbang, Abruzzi Katharine, Braun Rosemary
Department of Molecular Biosciences, Northwestern University, Evanston, Illinois, USA.
NSF-Simons Center for Quantitative Biology, Northwestern University, Evanston, Illinois, USA.
J Biol Rhythms. 2024 Dec;39(6):581-593. doi: 10.1177/07487304241273182. Epub 2024 Oct 8.
An autonomous, environmentally synchronizable circadian rhythm is a ubiquitous feature of life on Earth. In multicellular organisms, this rhythm is generated by a transcription-translation feedback loop present in nearly every cell that drives daily expression of thousands of genes in a tissue-dependent manner. Identifying the genes that are under circadian control can elucidate the mechanisms by which physiological processes are coordinated in multicellular organisms. Today, transcriptomic profiling at the single-cell level provides an unprecedented opportunity to understand the function of cell-level clocks. However, while many cycling detection algorithms have been developed to identify genes under circadian control in bulk transcriptomic data, it is not known how best to adapt these algorithms to single-cell RNA seq data. Here, we benchmark commonly used circadian detection methods on their reliability and efficiency when applied to single-cell RNA seq data. Our results provide guidance on adapting existing cycling detection methods to the single-cell domain and elucidate opportunities for more robust and efficient rhythm detection in single-cell data. We also propose a subsampling procedure combined with harmonic regression as an efficient strategy to detect circadian genes in the single-cell setting.
自主的、可与环境同步的昼夜节律是地球上生命的普遍特征。在多细胞生物中，这种节律由几乎每个细胞中存在的转录-翻译反馈环产生，该反馈环以组织依赖的方式驱动数千个基因的每日表达。识别受昼夜节律控制的基因可以阐明多细胞生物中生理过程的协调机制。如今，单细胞水平的转录组分析为理解细胞水平生物钟的功能提供了前所未有的机会。然而，虽然已经开发了许多循环检测算法来在大量转录组数据中识别受昼夜节律控制的基因，但尚不清楚如何最好地将这些算法应用于单细胞RNA测序数据。在这里，我们对常用的昼夜节律检测方法在应用于单细胞RNA测序数据时的可靠性和效率进行了基准测试。我们的结果为将现有的循环检测方法应用于单细胞领域提供了指导，并阐明了在单细胞数据中进行更稳健、更高效的节律检测的机会。我们还提出了一种结合谐波回归的子采样程序，作为在单细胞环境中检测昼夜节律基因的有效策略。
