SpatialExperiment: The SpatialExperiment class

In drighelli/VisiumExperiment: S4 Class for Spatially Resolved -omics Data

The SpatialExperiment class

Description

The SpatialExperiment class is designed to represent spatially resolved transcriptomics (ST) data. It inherits from the SingleCellExperiment class and is used in the same manner. In addition, the class supports storage of spatial information via spatialCoords and storage of images via imgData.

Arguments

...	Arguments passed to the SingleCellExperiment constructor to fill the slots of the base class.
sample_id	A character sample identifier, which matches the sample_id in imgData. The sample_id will also be stored in a new column in colData, if not already present. Default = sample01.
spatialCoordsNames	A character vector of column names from colData containing spatial coordinates, which will be accessible with spatialCoords. Alternatively, the spatialCoords argument may be provided. If both are provided, spatialCoordsNames is given precedence, and a warning is returned. Default = c("x", "y").
spatialCoords	A numeric matrix containing columns of spatial coordinates, which will be accessible with spatialCoords. Alternatively, spatialCoordsNames may be provided. If both are provided, spatialCoordsNames is given precedence, and a warning is returned.
scaleFactors	Optional scale factors associated with the image(s). This can be provided as a numeric value, numeric vector, list, or file path to a JSON file for the 10x Genomics Visium platform. For 10x Genomics Visium, the correct scale factor will automatically be selected depending on the resolution of the image from imageSources. Default = 1.
imgData	Optional DataFrame containing the image data. Alternatively, this can be built from the arguments imageSources and image_id (see Details).
imageSources	Optional file path(s) or URL(s) for one or more image sources.
image_id	Optional character vector (same length as imageSources) containing unique image identifiers.
loadImage	Logical indicating whether to load image into memory. Default = FALSE.
spatialDataNames	(Deprecated) A character vector of column names from colData to include in spatialData. Alternatively, the spatialData argument may be provided. If both are provided, spatialDataNames is given precedence, and a warning is returned. (Note: spatialData and spatialDataNames have been deprecated; colData and spatialCoords should be used for all columns. The arguments have been retained for backward compatibility but may be removed in the future.)
spatialData	(Deprecated) A DataFrame containing columns to store in spatialData, which must contain at least the columns of spatial coordinates. Alternatively, spatialDataNames may be provided. If both are provided, spatialDataNames is given precedence, and a warning is returned. (Note: spatialData and spatialDataNames have been deprecated; colData and spatialCoords should be used for all columns. The arguments have been retained for backward compatibility but may be removed in the future.)

Details

In this class, rows represent genes, and columns represent spots (for spot-based ST platforms) or cells (for molecule-based ST platforms). As for SingleCellExperiment, counts and logcounts can be stored in the assays slot, and row and column metadata in rowData and colData. For molecule-based ST data, the additional measurements per molecule per cell can be stored in a BumpyMatrix-formatted assay named molecules.

The additional arguments in the constructor documented above (e.g. spatialCoords, imgData, and others) represent the extensions to the SingleCellExperiment class to store associated spatial and imaging information for ST data.

The constructor expects colData to contain a column named sample_id. If this is not present, it will assign the value from the sample_id argument. If the imgData argument is provided, the constructor expects the imgData DataFrame to already be built. Otherwise, it will build it from the imageSources and (optional) image_id arguments. If image_id is not provided, this will be assumed from sample_id and imageSources instead. To combine multiple samples within a single object, see combine.

For 10x Genomics Visium datasets, the function read10xVisium can be used to load data and create a SpatialExperiment object directly from Space Ranger output files.

Value

A SpatialExperiment object.

Author(s)

Dario Righelli and Helena L. Crowell

See Also

?"SpatialExperiment-methods", which includes: spatialCoords, spatialCoordsNames, imgData, scaleFactors

?"SpatialExperiment-assays", which includes: molecules

?"SpatialExperiment-colData"

?"SpatialExperiment-combine"

?"SpatialExperiment-subset"

?"SpatialExperiment-misc"

readImgData

?"imgData-methods"

SpatialImage

read10xVisium

Examples

#########################################################
# Example 1: Spot-based ST (10x Visium) using constructor
#########################################################

dir <- system.file(
    file.path("extdata", "10xVisium", "section1", "outs"),
    package = "SpatialExperiment")

# read in counts
fnm <- file.path(dir, "raw_feature_bc_matrix")
sce <- DropletUtils::read10xCounts(fnm)

# read in image data
img <- readImgData(
    path = file.path(dir, "spatial"),
    sample_id="foo")

# read in spatial coordinates
fnm <- file.path(dir, "spatial", "tissue_positions_list.csv")
xyz <- read.csv(fnm, header = FALSE,
    col.names = c(
        "barcode", "in_tissue", "array_row", "array_col",
        "pxl_row_in_fullres", "pxl_col_in_fullres"))
    
# construct observation & feature metadata
rd <- S4Vectors::DataFrame(
    symbol = rowData(sce)$Symbol)
  
# construct 'SpatialExperiment'
(spe <- SpatialExperiment(
    assays = list(counts = assay(sce)),
    rowData = rd,
    colData = DataFrame(xyz),
    spatialCoordsNames = c("pxl_col_in_fullres", "pxl_row_in_fullres"),
    imgData = img,
    sample_id = "foo"))
    
#############################################################
# Example 2: Spot-based ST (10x Visium) using 'read10xVisium'
#############################################################

# see ?read10xVisium for details
example(read10xVisium)

##############################
# Example 3: Molecule-based ST
##############################

# create simulated data
n <- 1000; ng <- 50; nc <- 20
# sample xy-coordinates in [0,1]
x <- runif(n)
y <- runif(n)
# assign each molecule to some gene-cell pair
gs <- paste0("gene", seq(ng))
cs <- paste0("cell", seq(nc))
gene <- sample(gs, n, TRUE)
cell <- sample(cs, n, TRUE)
# construct data.frame of molecule coordinates
df <- data.frame(gene, cell, x, y)

# (assure gene & cell are factor so that
# missing observations aren't dropped)
df$gene <- factor(df$gene, gs)
df$cell <- factor(df$cell, cs)

# construct BumpyMatrix
mol <- BumpyMatrix::splitAsBumpyMatrix(
    df[, c("x", "y")],
    row = df$gene, column = df$cell)

# get count matrix
y <- with(df, table(gene, cell))
y <- as.matrix(unclass(y))

# construct SpatialExperiment
(spe_mol <- SpatialExperiment(
    assays = list(
        counts = y, 
        molecules = mol)))

drighelli/VisiumExperiment documentation built on April 10, 2024, 8:01 a.m.