This function performs unsupervised PCA on each mixscape class separately and projects each subspace onto all cells in the data. Finally, it uses the first 10 principle components from each projection as input to lda in MASS package together with mixscape class labels.

MixscapeLDA(
  object,
  assay = NULL,
  ndims.print = 1:5,
  nfeatures.print = 30,
  reduction.key = "LDA_",
  seed = 42,
  pc.assay = "PRTB",
  labels = "gene",
  nt.label = "NT",
  npcs = 10,
  verbose = TRUE,
  logfc.threshold = 0.25
)

Arguments

object

An object of class Seurat.

assay

Assay to use for performing Linear Discriminant Analysis (LDA).

ndims.print

Number of LDA dimensions to print.

nfeatures.print

Number of features to print for each LDA component.

reduction.key

Reduction key name.

seed

Value for random seed

pc.assay

Assay to use for running Principle components analysis.

labels

Meta data column with target gene class labels.

nt.label

Name of non-targeting cell class.

npcs

Number of principle components to use.

verbose

Print progress bar.

logfc.threshold

Limit testing to genes which show, on average, at least X-fold difference (log-scale) between the two groups of cells. Default is 0.25 Increasing logfc.threshold speeds up the function, but can miss weaker signals.

Value

Returns a Seurat object with LDA added in the reduction slot.