Context-explorer: Analysis of spatially organized protein expression in high-throughput screens

A growing body of evidence highlights the importance of the cellular microenvironment as a regulator of phenotypic and functional cellular responses to perturbations. We have previously developed cell patterning techniques to control population context parameters, and here we demonstrate context-explorer (CE), a software tool to improve investigation cell fate acquisitions through community level analyses. We demonstrate the capabilities of CE in the analysis of human and mouse pluripotent stem cells (hPSCs, mPSCs) patterned in colonies of defined geometries in multi-well plates. CE employs a density-based clustering algorithm to identify cell colonies. Using this automatic colony classification methodology, we reach accuracies comparable to manual colony counts in a fraction of the time, both in micropatterned and unpatterned wells. Classifying cells according to their relative position within a colony enables statistical analysis of spatial organization in protein expression within colonies. When applied to colonies of hPSCs, our analysis reveals a radial gradient in the expression of the transcription factors SOX2 and OCT4. We extend these analyses to colonies of different sizes and shapes and demonstrate how the metrics derived by CE can be used to asses the patterning fidelity of micropatterned plates. We have incorporated a number of features to enhance the usability and utility of CE. To appeal to a broad scientific community, all of the software’s functionality is accessible from a graphical user interface, and convenience functions for several common data operations are included. CE is compatible with existing image analysis programs such as CellProfiler and extends the analytical capabilities already provided by these tools. Taken together, CE facilitates investigation of spatially heterogeneous cell populations for fundamental research and drug development validation programs.

越来越多的研究证据表明，细胞微环境（cellular microenvironment）作为调控细胞对扰动产生表型与功能应答的关键因子，其重要性日益凸显。此前我们已开发细胞图案化技术以调控群体环境参数，本文中我们介绍了环境探索器（context-explorer, CE）——一款可通过群落水平分析推进细胞命运获取研究的软件工具。我们通过对多孔板内以指定几何形状构建群落的人源多能干细胞（hPSCs）与小鼠多能干细胞（mPSCs）开展分析，验证了CE的功能性能。CE采用基于密度的聚类算法（density-based clustering algorithm）识别细胞群落。借助这一自动化群落分类方法，无论在微图案化还是非图案化的多孔板孔中，我们均可在极短时间内获得与人工群落计数精度相当的结果。根据细胞在群落内的相对位置对其进行分类，可实现对群落内蛋白质表达空间组织特征的统计分析。当将该方法应用于人源多能干细胞群落时，我们的分析揭示了转录因子SOX2与OCT4的表达呈现径向梯度分布。我们将此类分析拓展至不同尺寸与形状的群落，并展示了CE所提取的量化指标如何用于评估微图案化板的图案保真度。我们为CE集成了多项特性以提升其易用性与实用价值。为覆盖更广泛的科研群体，该软件的所有功能均可通过图形用户界面（graphical user interface）调用，且内置了针对多种常见数据操作的便捷工具函数。CE可与现有图像分析工具（如CellProfiler）兼容，并可拓展此类工具已有的分析能力。综上，CE可助力针对空间异质性细胞群体的研究，适用于基础科研与药物开发验证项目。