Ewald Jessica D, Titterton Katherine L, Bäuerle Alex, Beatson Alex, Boiko Daniil A, Cabrera Ángel A, Cheah Jaime, Cimini Beth A, Gorissen Bram, Jones Thouis, Karczewski Konrad J, Rouquie David, Seal Srijit, Weisbart Erin, White Brandon, Carpenter Anne E, Singh Shantanu
Imaging Platform, Broad Institute of MIT and Harvard, Cambridge MA, USA.
Axiom Bio, San Francisco CA, USA.
bioRxiv. 2025 Jan 24:2025.01.22.634152. doi: 10.1101/2025.01.22.634152.
High-throughput, human-relevant approaches for predicting chemical toxicity are urgently needed for better decision-making in human health. Here, we apply image-based profiling (the Cell Painting assay) and two cytotoxicity assays (metabolic and membrane damage readouts) to primary human hepatocytes after exposure to eight concentrations of 1085 compounds that include pharmaceuticals, pesticides, and industrial chemicals with known liver toxicity-related outcomes. Three computational methods (CellProfiler, a Cell Painting-specific convolutional neural network, and a pretrained vision transformer) were compared to extract morphology features from single cells or entire images. We used these morphology features to predict activity in the measured cytotoxicity assays, as well as in 412 curated ToxCast assays that span cytotoxicity, cell-based, and cell-free categories. We found that the morphological profiles detect compound bioactivity at lower concentrations than standard cytotoxicity assays. In supervised analyses, they predict cytotoxicity and targeted cell-based assay readouts, but not cell-free assay readouts. We also found that the various feature extraction methods performed relatively similarly and that filtering out non-bioactive or cytotoxic concentrations did not boost supervised assay prediction performance for any assay endpoint category, although it did have a large influence on unsupervised cluster analysis. We envision that image-based profiling could serve as a key component of modern safety assessment.
为了在人类健康领域做出更好的决策，迫切需要高通量、与人类相关的化学毒性预测方法。在此，我们将基于图像的分析方法（细胞绘画分析）和两种细胞毒性分析方法（代谢和膜损伤读数）应用于原代人肝细胞，这些肝细胞在暴露于1085种化合物的八种浓度后，这些化合物包括具有已知肝脏毒性相关结果的药物、农药和工业化学品。比较了三种计算方法（CellProfiler、一种特定于细胞绘画的卷积神经网络和一种预训练的视觉变换器），以从单个细胞或整个图像中提取形态特征。我们使用这些形态特征来预测在测量的细胞毒性分析中以及在412种经过整理的ToxCast分析中的活性，这些分析涵盖细胞毒性、基于细胞和无细胞类别。我们发现，形态学分析在比标准细胞毒性分析更低的浓度下就能检测到化合物的生物活性。在监督分析中，它们可以预测细胞毒性和基于细胞的靶向分析读数，但不能预测无细胞分析读数。我们还发现，各种特征提取方法的表现相对相似，并且过滤掉无生物活性或细胞毒性的浓度并不会提高任何分析终点类别的监督分析预测性能，尽管它对无监督聚类分析有很大影响。我们设想基于图像的分析可以成为现代安全性评估的关键组成部分。
