Compute PCA on Integrated Multi-Slide Data

Performs PCA on the integrated data stored within the CoProMulti object. Assumes integration has created a common space across slides.

Usage

computePCA(
  object,
  nPCA = 40,
  center = TRUE,
  scale. = TRUE,
  scalePCs = TRUE,
  dataUse = "raw",
  center_per_slide = FALSE
)

Arguments

object

A CoProMulti object with the integratedData slot populated.

nPCA

Number of principal components to compute for each cell type.

center

Whether to center the matrix before PCA

scale.

Whether to scale the matrix before PCA

scalePCs

Whether to scale (whiten) PCs by their standard deviation before downstream CCA optimization. Default TRUE (recommended). When PCs are whitened, the unit-norm constraint ||w|| = 1 is equivalent to the standard CCA constraint ||w'X'Xw|| = 1. Setting this to FALSE distorts the constraint space and may produce unreliable results.

dataUse

What data to use, choices between "raw" and "integrated". Default is "raw". For single slide, this argument is ignored.

center_per_slide

After the global PCA, do we do center per slide again? By default this is set to FALSE

Value

A CoProMulti object with the pcaResults slot populated. pcaResults structure: list(slideID = list(cellType = pc_matrix)).

See also

computeDistance(), computeKernelMatrix(), runSkrCCA()

Other spatial-pipeline: computeDistance(), computeKernelMatrix(), runGeneSpaceCCA(), runSkrCCA()

Examples

toy <- readRDS(system.file("extdata", "toy_copro_data.rds", package = "CoPro"))
obj <- newCoProSingle(
  normalizedData = toy$normalizedData,
  locationData   = toy$locationData,
  metaData       = toy$metaData,
  cellTypes      = toy$cellTypes
)
obj <- subsetData(obj, cellTypesOfInterest = unique(toy$cellTypes))
obj <- computePCA(obj, nPCA = 10)
#> Input is dense (matrixarray), performing irlba pca...
#> Input is dense (matrixarray), performing irlba pca...