Synaptic dysfunction is a hallmark of various neurodegenerative and neurodevelopmental disorders. To interrogate synapse function in a systematic manner, we have established an automated high-throughput imaging pipeline based on fluorescence microscopy acquisition and image analysis of electrically stimulated synaptic transmission in neuronal cultures. Identification and measurement of synaptic signal fluctuations is achieved by means of an image analysis algorithm based on singular value decomposition. By exploiting the synchronicity of the evoked responses, the algorithm allows disentangling distinct temporally correlated patterns of firing synapse populations or cell types that are present in the same recording. We demonstrate the performance of the analysis with a pilot compound screen and show that the multiparametric readout allows classifying treatments by their spatiotemporal fingerprint. The image analysis and visualization software has been made publicly available on Github (https://www.github.com/S3Toolbox). The streamlined automation of multi-well image acquisition, electrical stimulation, analysis, and meta-data warehousing facilitates large-scale synapse-oriented screens and, in doing so, it will accelerate the drug discovery process.

突触功能障碍是各种神经退行性和神经发育障碍的标志。为了系统地询问突触功能，我们建立了一个基于荧光显微镜采集和神经元培养中电刺激突触传递的图像分析的自动化高通量成像管道。通过基于奇异值分解的图像分析算法实现突触信号波动的识别和测量。通过利用诱发反应的同步性，该算法允许解开存在于同一记录中的不同时间相关的触发突触群或细胞类型的模式。我们通过试验复合筛选展示了分析的性能，并表明多参数读数允许通过其时空指纹对治疗进行分类。图像分析和可视化软件已在 Github (https://www.github.com/S3Toolbox) 上公开。多孔图像采集、电刺激、分析和元数据仓库的简化自动化促进了大规模的面向突触的筛选，这样做将加速药物发现过程。