rliger (version 1.0.1)

optimizeSubset: Perform factorization for subset of data

Description

Uses an efficient strategy for updating that takes advantage of the information in the existing factorization. Can use either cell names or cluster names to subset. For more basic subsetting functionality (without automatic optimization), see subsetLiger.

Usage

optimizeSubset(
  object,
  cell.subset = NULL,
  cluster.subset = NULL,
  lambda = NULL,
  thresh = 1e-04,
  max.iters = 100,
  datasets.scale = NULL
)

Value

liger object with H, W, and V slots reset. Scale.data (if desired) will also be updated to reflect the subset.

Arguments

object

liger object. Should call optimizeALS before calling.

cell.subset

List of cell names to retain from each dataset (same length as number of datasets).

cluster.subset

Clusters for which to keep cells (ie. c(1, 5, 6)). Should pass in either cell.subset or cluster.subset but not both.

lambda

Regularization parameter. By default, uses last used lambda.

thresh

Convergence threshold. Convergence occurs when |obj0-obj|/(mean(obj0,obj)) < thresh (default 1e-4).

max.iters

Maximum number of block coordinate descent iterations to perform (default 100).

datasets.scale

Names of datasets to rescale after subsetting (default NULL).

Examples

ligerex <- createLiger(list(ctrl = ctrl, stim = stim))
ligerex <- normalize(ligerex)
ligerex <- selectGenes(ligerex)
ligerex <- scaleNotCenter(ligerex)
# \donttest{
# Assume we are performing the factorization
# Specification for minimal example run time, not converging.
ligerex <- optimizeALS(ligerex, k = 5, max.iters = 1)
# Preparing subset with random sampling.
# Subset can also be obtained with prior knowledge from metadata.
cell_names_1 <- sample(rownames(ligerex@H[[1]]), 20)
cell_names_2 <- sample(rownames(ligerex@H[[2]]), 20)

ligerex2 <- optimizeSubset(ligerex, cell.subset = list(cell_names_1, cell_names_2),
                           max.iters = 1)
# }