SMaSH

SMaSH identifies cell type-specific marker genes from single-cell RNA-sequencing (scRNA-seq) and spatial transcriptomics data to enable cell-type characterization and probe design.

Key Features:

• Scalability and Generalization: Handles large-scale scRNA-seq datasets with over 50 different cell types across hundreds of thousands of cells.
• Robust Marker Gene Extraction: Extracts robust and biologically relevant marker genes that characterize specific cellular populations, including niche phenotypes.
• Integration with ScanPy: Integrates with the ScanPy framework for scRNA-seq data analysis.
• Application to Spatial Transcriptomics: Evaluated on spatial transcriptomics data and capable of identifying highly localized cellular compartments such as those in the mouse cortex.

Scientific Applications:

• Differential Gene Expression Studies: Supports differential gene expression analyses across replicates, for example comparing tumour versus non-tumour environments.
• Cell Type Characterization and Validation: Aids precise characterization and validation of cell types within large datasets for disease-specific tissue analyses and human-wide cell atlases.
• Probe Design for Downstream Analyses: Provides marker genes usable for designing probes for subsequent experiments, including spatial transcriptomics.

Methodology:

SMaSH employs a general computational framework that emphasizes extraction of key marker genes tailored to specific cellular phenotypes, selecting markers that are statistically significant and biologically meaningful.