SpaCell

SpaCell integrates histopathology imaging and spatial transcriptomics (ST) data to analyze spatial relationships between tissue morphology and gene expression.

Key Features:

• Integration of Imaging and Gene Expression Data: Integrates millions of pixel intensity values with thousands of gene expression measurements from spatially-barcoded spots to link image-derived features with molecular profiles.
• Deep Learning-Based Analysis: Applies deep learning models to combined image and expression data to identify cell types and predict tissue labels.
• Improved Diagnostic Potential: Combines histopathological imaging with spatial sequencing data to enable quantitative, spatially informed disease characterization.

Scientific Applications:

• Histopathological Diagnosis: Provides spatial context that links morphology to gene expression to enhance disease diagnosis in histopathology.
• Discovery of Novel Biology: Enables discovery of spatially organized biological patterns by analyzing gene expression within tissue morphology.
• Cell Type Identification: Facilitates identification of cell types within tissue sections using integrated imaging and spatial gene expression signatures.

Methodology:

Integrates pixel intensity matrices with gene expression measurements from spatially-barcoded spots and trains deep learning models for cell type classification and tissue label prediction.