Tune nfeat using cross-validation.

CVTune_nfeat(
  reference,
  assayName = "logcounts",
  phenotypes = NULL,
  YY = NULL,
  ncomp = NULL,
  impScores = NULL,
  center = TRUE,
  scale = FALSE,
  nfeatLimits = c(10, 15000),
  gridSize = 30,
  nfeatV = NULL,
  Kfolds = 5,
  regMethod = "PCA",
  labelCode = "-1,1",
  seed = 5202056,
  Ncores = 1,
  nfeat = NULL
)

Arguments

reference: SingleCellExperiment object.
assayName: Character.
phenotypes: Vector.
YY: Matrix.
ncomp: Integer.
impScores: Matrix.
center: Logic.
scale: Logic.
nfeatLimits: Vector.
gridSize: Integer.
nfeatV: Vector.
Kfolds: Integer.
regMethod: Character.
labelCode: Character.
seed: Integer. The function sets the seed with withr::local_seed(), so the session's random-number state is restored on exit.
Ncores: Integer.
nfeat: Chosen number of features per phenotype, or NULL to choose after the cross-validation. Only an interactive session can choose: the function plots the loss and asks.

Value

A list.