Cell Decoder: decoding cell identity with multi-scale explainable deep learning

作者：Jun Zhu, Zeyang Zhang, Zeyang Zhang, Yujia Xiang, Beini Xie, Xinwen Dong, Linhai Xie, Peijie Zhou, Rongyan Yao, Li Yang, Yang Li, Fuchu He, Wenwu Zhu, Cheng Chang, Ziwei Zhang, Cheng Chang · 发表于：Genome biology · 年份：2025 · DOI：10.1186/s13059-025-03832-y · 被引用次数：4 · 研究领域：Single-cell and spatial transcriptomics、Cell Image Analysis Techniques、Machine Learning in Bioinformatics

BACKGROUND: Cells are the fundamental units of life, and understanding their diversity and functionality requires detailed characterization. The rise of single-cell omics data enables this, yet current deep learning approaches lack multi-scale interpretability. RESULTS: We introduce Cell Decoder, a model that integrates biological prior knowledge to provide a multi-scale representation of cells. Using automated machine learning and post hoc analysis, Cell Decoder decodes cell identity and outperforms existing methods. It offers multi-view interpretability and facilitates data integration. CONCLUSIONS: Applied to human bone and mouse embryonic data, Cell Decoder reveals the multi-scale heterogeneity of cell identities, providing a powerful framework for advancing our understanding of cellular diversity.