Researchers have developed mFISH3D, a sophisticated imaging technique that allows for the simultaneous visualization of multiple mRNA targets within intact, large-scale biological tissues. By combining fluorescence in situ hybridization with an AI-driven analysis platform called ZenCell, the workflow achieves precise single-cell mapping across entire organs like the mouse brain. This methodology overcomes historical limitations in depth and scalability, proving effective in diverse specimens including mouse embryos, kidneys, and even archived human brain tissue. The integration of self-supervised deep learning enables the automated identification of millions of cells with minimal manual effort. Consequently, this resource provides a robust framework for studying cellular diversity and gene expression patterns in both healthy and diseased states. This breakthrough facilitates a deeper understanding of spatial transcriptomics by maintaining the three-dimensional structural integrity of the analyzed organs.

References:

• Murakami T C, Xia M, Maeda Y, et al. Artificial Intelligence-driven Whole-brain Cell Mapping with Highly Multiplexed In Situ Hybridization[J]. bioRxiv, 2025: 2025.08. 18.670857.