Read an Expression Matrix and a Coordinate/Metadata Table
Source:R/readers_platform.R
ReadMatrixCoords.RdGeneral-purpose reader for the common spatial layout of one cell-by-feature
expression table plus one cell-metadata table holding coordinates. Each
argument may be a file path (read with data.table::fread), a matrix,
or a data frame.
Usage
ReadMatrixCoords(
expression,
metadata,
x_col = "x",
y_col = "y",
cell_id_col = NULL,
id_cols = character(0),
sample_id = "sample1",
transpose = FALSE,
project = "SpatialProject"
)Arguments
- expression
Cells-by-markers expression (matrix, data frame, or CSV path). Non-numeric and
id_colscolumns are dropped.- metadata
Cell metadata containing the coordinate columns (data frame or CSV path).
- x_col, y_col
Character. Coordinate column names in
metadata; values must be real numeric, finite vectors. Factor codes are never used as coordinate values.- cell_id_col
Character or
NULL. Shared cell-id column; if present in both tables, rows are matched on it, otherwise row order is assumed. Further columns (for example"fov") can follow: those present in both tables are matched as well, so ids need only be unique within them. The first column becomescell_id. Matched ids must be unique. Declared ID columns and reserved cell/sample identifiers in CSV files are read as text before type inference, so leading-zero identities remain distinct.- id_cols
Character vector. Non-marker id columns to drop from
expression(e.g."fov","cell_ID").- sample_id
Character. Sample identifier. Default
"sample1".- transpose
Logical. Set
TRUEifexpressionis markers-by-cells. DefaultFALSE.- project
Character. Project name.
Value
A SpatialCellData-class object.
Examples
expr <- matrix(rpois(40, 5), nrow = 10,
dimnames = list(NULL, c("CD3", "CD8", "CD20", "PanCK")))
meta <- data.frame(x = runif(10, 0, 100), y = runif(10, 0, 100))
obj <- ReadMatrixCoords(expr, meta)
obj
#> A SpatialCellData object
#> 10 cells across 1 sample
#> Markers: CD3, CD8, CD20, PanCK
#> Normalised: FALSE
#> Project: SpatialProject