Communication Direction Analysis

Infer spatial vector fields describing communication directions.

Compute spatial vector fields representing the average direction of cell-cell communication signals from senders to receivers.

Usage

communication_direction(
  seurat_obj,
  database_name,
  lr_pair = NULL,
  pathway_name = NULL,
  spatial_coords = NULL,
  k = 5L,
  bandwidth = NULL,
  kernel = c("exp", "lorentz"),
  normalize = TRUE,
  verbose = TRUE
)

Arguments

seurat_obj

Seurat object with COMMOTR results from spatial_communication.

database_name

Name of database used in spatial_communication.

lr_pair

Specific LR pair name (e.g., "Tgfb1-Tgfbr1_Tgfbr2"). If NULL, uses pathway_name.

pathway_name

Pathway name. Used if lr_pair is NULL. If both NULL, uses "total".

spatial_coords

Spatial coordinates: NULL (auto-detect), character (reduction name), or matrix.

k

Number of nearest neighbors for spatial smoothing (default: 5).

bandwidth

Bandwidth for kernel smoothing. If NULL, estimated from k.

kernel

Smoothing kernel: "exp" (exponential, default) or "lorentz".

normalize

Logical, normalize vectors to unit length (default: TRUE).

verbose

Logical, print progress messages (default: TRUE).

Value

Seurat object with vector field results stored in the COMMOTR results structure.

Details

For each cell, the function computes:

• Sender vector field: weighted average direction toward receiver cells

• Receiver vector field: weighted average direction from sender cells

The weights are communication matrix values smoothed with a spatial kernel. The resulting vector fields can be visualized using plot_cell_communication.

See also

spatial_communication, plot_cell_communication, communication_spatial_autocorrelation

Examples

if (FALSE) { # \dontrun{
# Compute direction for TGFb pathway
seurat_obj <- communication_direction(
    seurat_obj,
    database_name = "CellChat",
    pathway_name = "TGFb",
    k = 5
)

# Visualize
plot_cell_communication(seurat_obj, "CellChat", pathway_name = "TGFb")
} # }