Department of Biomedical Engineering, National University of Singapore, Singapore, 117411, Republic of Singapore.
Mechanobiology Institute, National University of Singapore, Singapore, 117411, Republic of Singapore.
Sci Rep. 2022 Sep 12;12(1):15329. doi: 10.1038/s41598-022-19472-2.
Cell morphology is profoundly influenced by cellular interactions with microenvironmental factors such as the extracellular matrix (ECM). Upon adhesion to specific ECM, various cell types are known to exhibit different but distinctive morphologies, suggesting that ECM-dependent cell morphological responses may harbour rich information on cellular signalling states. However, the inherent morphological complexity of cellular and subcellular structures has posed an ongoing challenge for automated quantitative analysis. Since multi-channel fluorescence microscopy provides robust molecular specificity important for the biological interpretations of observed cellular architecture, here we develop a deep learning-based analysis pipeline for the classification of cell morphometric phenotypes from multi-channel fluorescence micrographs, termed SE-RNN (residual neural network with squeeze-and-excite blocks). We demonstrate SERNN-based classification of distinct morphological signatures observed when fibroblasts or epithelial cells are presented with different ECM. Our results underscore how cell shapes are non-random and established the framework for classifying cell shapes into distinct morphological signature in a cell-type and ECM-specific manner.
细胞形态受到与细胞外基质（ECM）等微环境因素相互作用的深刻影响。已知当特定的细胞附着于 ECM 时，各种细胞类型会表现出不同但独特的形态，这表明 ECM 依赖性的细胞形态响应可能蕴藏着丰富的细胞信号状态信息。然而，细胞和亚细胞结构固有的形态复杂性一直是自动化定量分析的挑战。由于多通道荧光显微镜为观察到的细胞结构的生物学解释提供了重要的稳健分子特异性，因此我们开发了一种基于深度学习的分析管道，用于从多通道荧光显微图像中分类细胞形态计量表型，称为 SE-RNN（带有挤压和激发块的残差神经网络）。我们证明了当成纤维细胞或上皮细胞呈现不同的 ECM 时，可以对观察到的不同形态特征进行基于 SERNN 的分类。我们的结果强调了细胞形状是非随机的，并为以细胞类型和 ECM 特异性的方式将细胞形状分类为不同的形态特征建立了框架。
