Peñas Kristofer E Delas, Haeusler Ralf, Feng Sally, Magidson Valentin, Dmitrieva Mariia, Wink David, Lockett Stephen, Kinders Robert, Rittscher Jens
Department of Engineering Science, University of Oxford, United Kingdom.
Big Data Institute, University of Oxford, Li Ka Shing Centre for Health Information and Discovery, Oxford, UK.
Med Opt Imaging Virtual Microsc Image Anal (2022). 2022 Sep 15:84-93. doi: 10.1007/978-3-031-16961-8_9.
The morphology of individual cells can reveal much about the underlying states and mechanisms in biology. In tumor environments, the interplay among different cell morphologies in local neighborhoods can further improve this characterization. In this paper, we present an approach based on representation learning to capture similarities and subtle differences in cells positive for H2AX, a common marker for DNA damage. We demonstrate that texture representations using GLCM and VAE-GAN enable profiling of cells in both singular and local neighborhood contexts. Additionally, we investigate a possible quantification of immune and DNA damage response interplay by enumerating CD8+ and H2AX+ on different scales. Using our profiling approach, regions in treated tissues can be differentiated from control tissue regions, demonstrating its potential in aiding quantitative measurements of DNA damage and repair in tumor contexts.
单个细胞的形态可以揭示许多关于生物学潜在状态和机制的信息。在肿瘤环境中，局部区域内不同细胞形态之间的相互作用可以进一步完善这种特征描述。在本文中，我们提出了一种基于表征学习的方法，以捕捉H2AX（一种常见的DNA损伤标记物）阳性细胞中的相似性和细微差异。我们证明，使用灰度共生矩阵（GLCM）和变分自编码器-生成对抗网络（VAE-GAN）的纹理表征能够在单个细胞和局部区域环境中对细胞进行分析。此外，我们通过在不同尺度上枚举CD8+和H2AX+来研究免疫与DNA损伤反应相互作用的一种可能的量化方法。使用我们的分析方法，可以将处理过的组织区域与对照组织区域区分开来，证明了其在辅助肿瘤环境中DNA损伤和修复的定量测量方面的潜力。
