Package 'sosta'

Title: A package for the analysis of anatomical tissue structures in spatial omics data
Description: `sosta` (Spatial Omics STructure Analysis) is a package for analyzing spatial omics data to explore tissue organization at the anatomical structure level. It reconstructs morphologically relevant structures based on molecular features or cell types. It further calculates a range of structural and shape metrics to quantitatively describe tissue architecture. The package is designed to integrate with other packages for the analysis of spatial (omics) data.
Authors: Samuel Gunz [aut, cre] , Mark D. Robinson [aut, fnd]
Maintainer: Samuel Gunz <[email protected]>
License: GPL (>= 3) + file LICENSE
Version: 0.99.3
Built: 2025-01-20 03:29:34 UTC
Source: https://github.com/bioc/sosta

Help Index

Function to estimate the intensity image of a point pattern

Description

Function to estimate the intensity image of a point pattern

Usage

.intensityImage(ppp, mark_select = NULL, bndw = NULL, dim)

Arguments

ppp

point pattern object of class ppp
mark_select

character; name of mark that is to be selected for the reconstruction
bndw

bandwidth of kernel density estimator
dim

numeric; x dimension of the final reconstruction.

Value

list; list with the intensity image and the bandwidth and dimension parameters

Function to estimate the intensity threshold for the reconstruction of spatial structures

Description

Function to estimate the intensity threshold for the reconstruction of spatial structures

Usage

.intensityThreshold(density_image, steps = 250)

Arguments

density_image

real-valued pixel image; output from the function .intensityImage
steps

numeric; value used to filter the density estimates, where only densities greater than the maximum value divided by threshold are considered. Default is 250.

Value

numeric; estimated threshold

Converts a binary matrix to an sf polygon

Description

Converts a binary matrix to an sf polygon

Usage

binaryImageToSF(binaryMatrix, xmin, xmax, ymin, ymax)

Arguments

binaryMatrix

matrix; binary matrix
xmin

integer; minimum x coordinate of the coordinate system
xmax

integer; maximum x coordinate of the coordinate system
ymin

integer; minimum y coordinate of the coordinate system
ymax

integer; maximum y coordinate of the coordinate system

Value

sf object

Examples

matrix_R <- matrix(c(
    0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0
), nrow = 9, byrow = TRUE)
poly_R <- binaryImageToSF(matrix_R, xmin = 0, xmax = 1, ymin = 0, ymax = 1)
plot(poly_R)

Estimate reconstruction parameters from a set of images

Description

Estimate reconstruction parameters from a set of images

Usage

estimateReconstructionParametersSPE(
  spe,
  marks,
  image_col,
  mark_select = NULL,
  nimages = NULL,
  fun = "bw.diggle",
  dim = 500,
  ncores = 1,
  plot_hist = TRUE
)

Arguments

spe

SpatialExperiment; a object of class SpatialExperiment
marks

character; name of column in colData that will correspond to the ppp marks
image_col

character; name of a column in colData that corresponds to the image
mark_select

character; name of mark that is to be selected for the reconstruction
nimages

integer; number of images for the estimation. Will be randomly sampled
fun

character; function to estimate the kernel density. Default bw.diggle.
dim

numeric; x dimension of the final reconstruction. A lower resolution speed up computation but lead to less exact reconstruction. Default = 500
ncores

numeric; number of cores for parallel processing using mclapply. Default = 1
plot_hist

logical; if histogram of estimated densities and thresholds should be plotted. Default = TRUE

Value

tibble; tibble with estimated intensities and thresholds

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
spe_sel <- spe[, spe[["image_name"]] %in% c("E02", "E03", "E04")]
estimateReconstructionParametersSPE(spe_sel,
    marks = "cell_category",
    image_col = "image_name", mark_select = "islet", plot_hist = TRUE
)

Estimate the intensity threshold for the reconstruction of spatial structures

Description

Estimate the intensity threshold for the reconstruction of spatial structures

Usage

findIntensityThreshold(ppp, mark_select = NULL, bndw = NULL, dim, steps = 250)

Arguments

ppp

point pattern object of class ppp
mark_select

character; name of mark that is to be selected for the reconstruction
bndw

numeric; bandwith of the sigma parameter in the density estimation, if no value is given the bandwith is estimated using cross validation with the bw.diggle function.
dim

numeric; x dimension of the final reconstruction.
steps

numeric; value used to filter the density estimates, where only densities greater than the maximum value divided by threshold are considered. Default is 250.

Value

numeric; estimated intensity threshold

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
ppp <- SPE2ppp(spe, marks = "cell_category", image_col = "image_name", image_id = "E04")
findIntensityThreshold(ppp, mark_select = "islet", dim = 250)

Function to get the dimension based on dim of y axis

Description

Function to get the dimension based on dim of y axis

Usage

getDimXY(ppp, ydim)

Arguments

ppp

point pattern object of class ppp
ydim

dimension of y axis

Value

vector; vector with x and y dimension

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
ppp <- SPE2ppp(spe,
    marks = "cell_category", image_col = "image_name",
    image_id = "E04"
)
getDimXY(ppp, 500)

Calculate mean shape metrics of a set of polygons

Description

Calculate mean shape metrics of a set of polygons

Usage

meanShapeMetrics(totalShapeMetricMatrix)

Arguments

totalShapeMetricMatrix

matrix of shape metrics

Value

matrix; matrix of mean shape metrics

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
islet_poly <- reconstructShapeDensityImage(spe,
    marks = "cell_category",
    image_col = "image_name", image_id = "E04", mark_select = "islet", dim = 500
)
shape_metrics <- totalShapeMetrics(islet_poly)
meanShapeMetrics(shape_metrics)

Function to normalize coordinates between zero and one while keep scaling

Description

Function to normalize coordinates between zero and one while keep scaling

Usage

normalizeCoordinates(coords)

Arguments

coords

matrix; matrix with coordinates

Value

matrix; coordinates scaled between 0 and 1

Examples

matrix_R <- matrix(c(
    0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0
), nrow = 9, byrow = TRUE)
coords <- xyCoordinates(matrix_R)
normalizeCoordinates(coords)

Reconstruct polygon from point pattern density

Description

This function estimates the density of a spatial point pattern (ppp), thresholds the density to create a binary image, and then converts it to a valid sf object (polygons).

Usage

reconstructShapeDensity(
  ppp,
  mark_select = NULL,
  bndw = NULL,
  thres = NULL,
  dim
)

Arguments

ppp

point pattern object of class ppp
mark_select

character; name of mark that is to be selected for the reconstruction
bndw

bandwidth of kernel density estimator
thres

intensity threshold for the reconstruction
dim

numeric; x dimension of the final reconstruction.

Value

sf object of class POLYGON

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
ppp <- SPE2ppp(spe, marks = "cell_category", image_col = "image_name", image_id = "E04")
thres <- findIntensityThreshold(ppp, mark_select = "islet", dim = 500)
islet_poly <- reconstructShapeDensity(ppp, mark_select = "islet", thres = thres, dim = 500)
plot(islet_poly)

Reconstruct structure from spe object with given image id

Description

Reconstruct structure from spe object with given image id

Usage

reconstructShapeDensityImage(
  spe,
  marks,
  image_col,
  image_id,
  mark_select,
  dim = 500,
  bndw = NULL,
  thres = NULL
)

Arguments

spe

SpatialExperiment; a object of class SpatialExperiment
marks

character; name of column in colData that will correspond to the ppp marks
image_col

character; name of a column in colData that corresponds to the image
image_id

character; image id, must be present in image_col
mark_select

character; name of mark that is to be selected for the reconstruction
dim

numeric; x dimension of the final reconstruction. A lower resolution speed up computation but lead to less exact reconstruction. Default = 500
bndw

numeric; bandwith of the sigma parameter in the density estimation, if no value is given the bandwith is estimated using cross validation with the bw.diggle function.
thres

numeric; intensity threshold for the reconstruction

Value

sf object of class POLYGON

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
islet_poly <- reconstructShapeDensityImage(spe,
    marks = "cell_category",
    image_col = "image_name", image_id = "E04", mark_select = "islet", dim = 500
)
plot(islet_poly)

Reconstruct structure from spatial experiment object per image id

Description

Reconstruct structure from spatial experiment object per image id

Usage

reconstructShapeDensitySPE(
  spe,
  marks,
  image_col,
  mark_select,
  dim = 500,
  bndw = NULL,
  thres,
  ncores = 1
)

Arguments

spe

SpatialExperiment; a object of class SpatialExperiment
marks

character; name of column in colData that will correspond to the ppp marks
image_col

character; name of a column in colData that corresponds to the image
mark_select

character; name of mark that is to be selected for the reconstruction
dim

numeric; x dimension of the final reconstruction. A lower resolution speed up computation but lead to less exact reconstruction. Default = 500
bndw

numeric; bandwith of the sigma parameter in the density estimation, if no value is given the bandwith is estimated using cross validation with the bw.diggle function.
thres

numeric; intensity threshold for the reconstruction
ncores

numeric; number of cores for parallel processing using mclapply. Default = 1

Value

simple feature collection

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
spe_sel <- spe[, spe[["image_name"]] %in% c("E02", "E03", "E04")]
all_islets <- reconstructShapeDensitySPE(spe_sel,
    marks = "cell_category",
    image_col = "image_name", mark_select = "islet", bndw = sigma, thres = 0.0025
)
all_islets

Intensity plot

Description

This function plots the intensity of a point pattern image and displays a histogram of the intensity values. Note that intensities less than largest intensity value divided by 250 are not displayed in the histogram.

Usage

shapeIntensityImage(
  spe,
  marks,
  image_col,
  image_id,
  mark_select,
  bndw = NULL,
  dim = 500
)

Arguments

spe

SpatialExperiment; a object of class SpatialExperiment
marks

character; name of column in colData that will correspond to the ppp marks
image_col

character; name of a column in colData that corresponds to the image
image_id

character; image id, must be present in image_col
mark_select

character; name of mark that is to be selected for the reconstruction
bndw

numeric; bandwith of the sigma parameter in the density estimation, if no value is given the bandwith is estimated using cross validation with the bw.diggle function.
dim

numeric; x dimension of the final reconstruction. A lower resolution speeds up computation but lead to less exact reconstruction. Default = 500

Value

ggplot object with intensity image and histogram

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
shapeIntensityImage(spe,
    marks = "cell_category", image_col = "image_name",
    image_id = "E04", mark_select = "islet"
)

Calculate a set of shape metrics of a polygon

Description

Calculate a set of shape metrics of a polygon

Usage

shapeMetrics(sfPoly)

Arguments

sfPoly

POLYGON of class sfc

Value

list; list of shape metrics

Examples

matrix_R <- matrix(c(
    0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0
), nrow = 9, byrow = TRUE)
poly_R <- binaryImageToSF(matrix_R, xmin = 0, xmax = 1, ymin = 0, ymax = 1)
shapeMetrics(poly_R)

Function to convert spatial coordinates of a SpatialExperiment object to a ppp object

Description

Function to convert spatial coordinates of a SpatialExperiment object to a ppp object

Usage

SPE2ppp(spe, marks, image_col = NULL, image_id = NULL)

Arguments

spe

SpatialExperiment; a object of class SpatialExperiment
marks

character; name of column in colData that will correspond to the ppp marks
image_col

character; name of a column in colData that corresponds to the image
image_id

character; image id, must be present in image_col

Value

ppp; object of type ppp

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
SPE2ppp(spe, marks = "cell_category", image_col = "image_name", image_id = "E04")

Title

Description

Title

Usage

st_calculateCurvature(sfPoly, smoothness = 5)

Arguments

sfPoly

POLYGON of class sf
smoothness

list; curvature measures

Value

list; list of curvatures values

References

https://stackoverflow.com/questions/62250151/calculate-curvature-of-a-closed-object-in-r

Examples

matrix_R <- matrix(c(
    0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0
), nrow = 9, byrow = TRUE)
poly_R <- binaryImageToSF(matrix_R, xmin = 0, xmax = 1, ymin = 0, ymax = 1)
st_calculateCurvature(poly_R)

Calculate curl of a polygon

Description

Calculate curl of a polygon

Usage

st_calculateShapeCurl(sfPoly)

Arguments

sfPoly

POLYGON of class sf

Value

numeric; the curl of the polygon

Examples

matrix_R <- matrix(c(
    1, 1, 1, 1, 1, 0,
    1, 1, 0, 0, 1, 1,
    1, 1, 0, 0, 1, 1,
    1, 1, 1, 1, 1, 0,
    1, 1, 0, 1, 1, 0,
    1, 1, 0, 0, 1, 1,
    1, 1, 0, 0, 1, 1
), nrow = 7, byrow = TRUE)
poly_R <- binaryImageToSF(matrix_R, xmin = 0, xmax = 1, ymin = 0, ymax = 1)
st_calculateShapeCurl(poly_R)

Calculate the length of feature axes of an sf polygon

Description

Calculate the length of feature axes of an sf polygon

Usage

st_feature_axes(sfPoly)

Arguments

sfPoly

POLYGON of class sf

Value

list; list containing the major and minor axis lengths

Examples

matrix_R <- matrix(c(
    0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0
), nrow = 9, byrow = TRUE)
poly_R <- binaryImageToSF(matrix_R, xmin = 0, xmax = 1, ymin = 0, ymax = 1)
st_feature_axes(poly_R)

Calculate a set of shape metrics of a set of polygons

Description

Calculate a set of shape metrics of a set of polygons

Usage

totalShapeMetrics(sfInput)

Arguments

sfInput

MULTIPOLYGON of class sf

Details

Calculate a set of shape metrics of a set of polygons. The function calculates all metrics that are implemented in the function shapeMetrics()

Value

matrix; matrix of shape metrics

Examples

spe <- imcdatasets::Damond_2019_Pancreas("spe", full_dataset = FALSE)
islet_poly <- reconstructShapeDensityImage(spe,
    marks = "cell_category",
    image_col = "image_name", image_id = "E04", mark_select = "islet", dim = 500
)
totalShapeMetrics(islet_poly)

Function to extract x y coordinates from binary image

Description

Function to extract x y coordinates from binary image

Usage

xyCoordinates(inputMatrix)

Arguments

inputMatrix

a binary matrix

Value

matrix; matrix with x,y coordinates of the cell of the input matrix

Examples

matrix_R <- matrix(c(
    0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 1, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 1, 1, 0, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 1, 1, 0, 0, 1, 1, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0
), nrow = 9, byrow = TRUE)
xyCoordinates(matrix_R)