Skip to contents

Calculate Spatial Organization Entropy (OSE) for samples

Source: R/MultiSampAnalysis.R
CalSampOSE.Rd

Performs spatial organization entropy analysis to quantify tissue spatial heterogeneity based on cell type distribution patterns. This function implements the OSE algorithm which partitions tissue into regions and calculates entropy based on local cell type diversity.

Usage

CalSampOSE(
  STID_obj = NULL,
  loop_id = NULL,
  meta_key = NULL,
  group_by = NULL,
  OSE_dist_m = 0.2,
  OSE_PCA_nPC = 15,
  OSE_window = NULL,
  OSE_minSpotNum = NULL,
  only_plot = FALSE,
  plot_params = list(p3_size = 2, p4_size = 0.3, p2_merge_r = 15, p3_merge_size = 9),
  col = COLOR_LIST[["PALETTE_WHITE_BG"]],
  return_data = FALSE,
  grp_nm = NULL,
  dir_nm = "M4_CalSampOSE"
)

Arguments

STID_obj

A STID object containing spatial transcriptomics data

loop_id

Character, multi-sample analysis identifier

meta_key

Character, metadata key for retrieving cell data

group_by

Character, column name for cell type grouping

OSE_dist_m

Numeric, distance weight (default: 0.2). Smaller values give more weight to expression distance in region segmentation

OSE_PCA_nPC

Integer, number of PCs for PCA dimensionality reduction (default: 15)

OSE_window

Numeric, window size for superpixel generation. If NULL, automatically set to max(x_range, y_range)/10

OSE_minSpotNum

Integer, minimum spots per region. If NULL, automatically set to nrow(meta_data)/1000

only_plot

Logical, whether to only regenerate plots from existing results (default: FALSE)

plot_params

List of plotting parameters:

  • p3_size: Point size for partition plot

  • p4_size: Point size for group plot

  • p2_merge_r: Radius for pie chart in merged plot

  • p3_merge_size: Point size for entropy in merged plot

col

Character vector, color palette for cell types

return_data

Logical, whether to return results as a list (default: FALSE)

grp_nm

Character, group name for output directory (default: NULL, uses timestamp)

dir_nm

Character, directory name for output (default: "M4_CalSampOSE")

Value

If return_data = TRUE, returns a list containing:

  • center_df: Data frame of region centers

  • data: Cell-level OSE data with region assignments

  • entropy_results: Region entropy calculations

  • celltype_percent: Cell type proportions per region

  • segment_df: Region boundary segments

Examples

if (FALSE) { # \dontrun{
results <- CalSampOSE(
  STID_obj = stid_object,
  loop_id = "comparison_1",
  meta_key = "M2_NicheDetect_STS_20240101",
  group_by = "cell_type",
  OSE_window = 50,
  OSE_minSpotNum = 100
)
} # }