BOMS

BOMS segments cells in spatial transcriptomics using a mean-shift approach that incorporates spatial location and gene labels of mRNA spots without requiring auxiliary images.

Key Features:

• Mean-shift segmentation: Segments cells based on the spatial locations and gene labels of mRNA spots using a mean-shift algorithm.
• No auxiliary image requirement: Operates without DAPI, Poly(A), or other auxiliary stains, avoiding difficulties with stain segmentation and manual parameter tuning.
• Computational efficiency: Achieves comparable results to existing methods while being significantly faster in execution.
• Simple implementation: Provides an open-source implementation that is straightforward to deploy.

Scientific Applications:

• Cell segmentation in spatial transcriptomics: Assigns mRNA molecules to individual cells in imaging-based spatial transcriptomics data at subcellular resolution.
• Gene expression analysis: Reduces misassignment errors in downstream analysis of gene expression and regulation in complex tissues.
• Benchmarking across datasets: Evaluated on various publicly available spatial transcriptomics datasets.

Methodology:

BOMS applies mean shift to segment cells using the spatial locations and gene labels of mRNA spots.