RunSDA: RunSDA

In bimberlabinternal/CellMembrane: A package with high-level wrappers and pipelines for single-cell RNA-seq tools

RunSDA

Description

This will run SDA on the target assay

Usage

RunSDA(
  seuratObj,
  outputFolder,
  numComps = 50,
  minCellsExpressingFeature = 0.01,
  perCellExpressionThreshold = 2,
  minFeatureCount = 1,
  featureInclusionList = NULL,
  featureExclusionList = NULL,
  maxFeaturesDiscarded = NULL,
  assayName = "RNA",
  randomSeed = GetSeed(),
  minLibrarySize = 50,
  path.sda = "sda_static_linux",
  max_iter = 10000,
  save_freq = 1000,
  nThreads = 1,
  storeGoEnrichment = FALSE,
  sdaDebug = FALSE
)

Arguments

seuratObj	A Seurat object.
outputFolder	The path to save results. There will be subfolders for ./rawData and ./results
numComps	Passed to SDAtools::run_SDA(). 30 is a good minimum but depends on input data complexity.
minCellsExpressingFeature	Can be used with perCellExpressionThreshold to drop features present in limited cells. Only features expressed >= perCellExpressionThreshold in at least minCellsExpressingFeature cells will be retained. If this value is less than one it is interpreted as a percentage of total cells. If above one it is interpreted as the min number of cells.
perCellExpressionThreshold	Can be used with perCellExpressionThreshold to drop features present in limited cells. Only features expressed >= perCellExpressionThreshold in at least minCellsExpressingFeature cells will be retained.
minFeatureCount	Only features where the total counts across all cells are greater than or equal to this value will be included. Setting this value to one will include all expressed genes.
featureInclusionList	An optional vector of genes that will be included in SDA
featureExclusionList	An optional vector of genes that will be excluded from SDA
maxFeaturesDiscarded	After filtering, if fewer than this number of features remain, an error will be thrown. This is a guard against overly aggressive filtering. This can either be an integer (meaning number of features), or 0-1 (which is interpreted as a percent of the input features)
assayName	The name of the assay
randomSeed	Passed to SDAtools::run_SDA() set_seed
minLibrarySize	Passed to dropsim::normaliseDGE() min_library_size. Only cells with library size equal or greater to this will be kept. IMPORTANT: this is applied after feature selection.
path.sda	The full path to the SDA binary. By default it assumes sda_static_linux in in your $PATH
max_iter	Passed directly to SDAtools::run_SDA()
save_freq	Passed directly to SDAtools::run_SDA()
nThreads	Passed to SDAtools::run_SDA() num_openmp_threads
storeGoEnrichment	If true, SDA_GO_Enrichment will be performed and stored in the result list
sdaDebug	This is passed directly to SDAtools::run_SDA()

bimberlabinternal/CellMembrane documentation built on Dec. 25, 2024, 8:10 a.m.