Banksy

Banksy incorporates spatial information to cluster cells in a feature space such as gene expression, enabling spatially-aware cell typing and tissue domain segmentation.

Key Features:

• Mean neighborhood expression computation: Banksy computes mean neighborhood expression to capture local spatial context around each cell.
• Azimuthal Gabor filters: Banksy computes azimuthal Gabor filters that capture gene expression gradients in the spatial neighborhood.
• Neighbor-augmented product space embedding: Banksy combines spatial features with the cell's own expression to embed cells in a neighbor-augmented product space for subsequent clustering.

Scientific Applications:

• Spatially-aware cell typing: Banksy enables accurate cell typing that accounts for spatial context in transcriptomic data.
• Tissue domain segmentation: Banksy supports segmentation of tissue domains based on spatial and expression features.

Methodology:

Banksy computes mean neighborhood expression and azimuthal Gabor filters that capture gene expression gradients, then combines these spatial features with the cell's own expression to embed cells in a neighbor-augmented product space which is then clustered.