Shi Yu, Tabet Jimmy S, Milkie Daniel E, Daugird Timothy A, Yang Chelsea Q, Giovannucci Andrea, Legant Wesley R
bioRxiv. 2023 Mar 9:2023.03.07.531517. doi: 10.1101/2023.03.07.531517.
Light sheet microscopes enable rapid, high-resolution imaging of biological specimens; however, biological processes span a variety of spatiotemporal scales. Moreover, long-term phenotypes are often instigated by rare or fleeting biological events that are difficult to capture with a single imaging modality and constant imaging parameters. To overcome this limitation, we present smartLLSM, a microscope that incorporates AI-based instrument control to autonomously switch between epifluorescent inverted imaging and lattice light sheet microscopy. We apply this technology to two major scenarios. First, we demonstrate that the instrument provides population-level statistics of cell cycle states across thousands of cells on a coverslip. Second, we show that by using real-time image feedback to switch between imaging modes, the instrument autonomously captures multicolor 3D datasets or 4D time-lapse movies of dividing cells at rates that dramatically exceed human capabilities. Quantitative image analysis on high-content + high-throughput datasets reveal kinetochore and chromosome dynamics in dividing cells and determine the effects of drug perturbation on cells in specific mitotic stages. This new methodology enables efficient detection of rare events within a heterogeneous cell population and records these processes with high spatiotemporal 4D imaging over statistically significant replicates.
光片显微镜能够对生物样本进行快速、高分辨率成像;然而,生物过程跨越了各种时空尺度。此外,长期表型往往是由罕见或短暂的生物事件引发的,这些事件很难用单一成像方式和固定成像参数捕捉到。为了克服这一限制,我们推出了smartLLSM,这是一种结合了基于人工智能的仪器控制的显微镜,能够在落射荧光倒置成像和晶格光片显微镜之间自动切换。我们将这项技术应用于两个主要场景。首先,我们证明该仪器能够提供盖玻片上数千个细胞的细胞周期状态的群体水平统计数据。其次,我们表明,通过使用实时图像反馈在成像模式之间切换,该仪器能够以远远超过人类能力的速度自动捕获分裂细胞的多色3D数据集或4D延时电影。对高内涵+高通量数据集的定量图像分析揭示了分裂细胞中的动粒和染色体动态,并确定了药物扰动对处于特定有丝分裂阶段的细胞的影响。这种新方法能够在异质细胞群体中高效检测罕见事件,并通过具有统计学意义的重复实验,以高时空4D成像记录这些过程。