subsetLigerDataset: Subset ligerDataset object

In rliger: Linked Inference of Genomic Experimental Relationships

Subset ligerDataset object

Description

This function subsets a ligerDataset object with valid feature and cell indices. For HDF5 based object, options are available for subsetting data into memory or a new on-disk H5 file. Feature and cell subscription is always based on the size of rawData. Therefore, the feature subsetting on scaled data, which usually contains already a subset of features, will select the intersection between the wanted features and the set available from scaled data.

Usage

subsetLigerDataset(
  object,
  featureIdx = NULL,
  cellIdx = NULL,
  useSlot = NULL,
  newH5 = TRUE,
  filename = NULL,
  filenameSuffix = NULL,
  chunkSize = 1000,
  verbose = getOption("ligerVerbose", TRUE),
  returnObject = TRUE,
  ...
)

subsetH5LigerDataset(
  object,
  featureIdx = NULL,
  cellIdx = NULL,
  useSlot = NULL,
  newH5 = TRUE,
  filename = NULL,
  filenameSuffix = NULL,
  chunkSize = 1000,
  verbose = getOption("ligerVerbose", TRUE),
  returnObject = TRUE
)

subsetMemLigerDataset(
  object,
  featureIdx = NULL,
  cellIdx = NULL,
  useSlot = NULL,
  returnObject = TRUE
)

Arguments

object	ligerDataset object. HDF5 based object if using subsetH5LigerDataset, in-memory data for subsetMemLigerDataset.
featureIdx	Character, logical or numeric index that can subscribe features. Missing or NULL for all features.
cellIdx	Character, logical or numeric index that can subscribe cells. Missing or NULL for all cells.
useSlot	The slot(s) to only consider. Choose one or more from "rawData", "normData" and "scaleData". Default NULL subsets the whole object including analysis result matrices.
newH5	Whether to create a new H5 file on disk for the subset dataset if object is HDF5 based. TRUE writes a new one, FALSE returns in memory data.
filename	Filename of the new H5 file if being created. Default NULL adds suffix ".subset_{yymmdd_HHMMSS}.h5" to the original name.
filenameSuffix	Instead of specifying the exact filename, set a suffix for the new files so the new filename looks like original.h5.[suffix].h5. Default NULL.
chunkSize	Integer. Number of maximum number of cells in each chunk, Default 1000.
verbose	Logical. Whether to show information of the progress. Default getOption("ligerVerbose") or TRUE if users have not set.
returnObject	Logical, whether to return a ligerDataset object for result. Default TRUE. FALSE returns a list containing requested values.
...	Arguments passed to subsetH5LigerDataset

Value

Subset object

Examples

ctrl <- dataset(pbmc, "ctrl")
ctrl.small <- subsetLigerDataset(ctrl, cellIdx = 1:5)
ctrl.tiny <- ctrl[1:5, 1:5]

rliger documentation built on Oct. 30, 2024, 1:07 a.m.