CellART: a unified framework for extracting single-cell information from high-resolution spatial transcriptomics.

CellART accurately extracts comprehensive single-cell information from high-resolution spatial transcriptomics data using multimodal integration.

• Why it matters: Current spatial transcriptomics platforms face limitations in transcript count density and gene coverage, hindering detailed single-cell analysis. Addressing these gaps can improve understanding of cellular heterogeneity and spatial organization.
• What they did: The framework combines staining images, spatial transcriptomics data, and single-cell RNA sequencing references, employing deep learning and probabilistic modeling for simultaneous cell segmentation and cell-type annotation.
• The result: CellART is efficient, generalizable, and robust, producing outputs compatible with community tools, enabling diverse downstream analyses of spatial transcriptomics data.