gecd_CellDistinguisher: The gecd_CellDistinguisher function

In GeneralElectric/CellDistinguisher: Computational de novo Discovery of Distinguishing Genes for Biological Processes and Cell Types in Complex Tissues

Description

Finds distinguishers that identify specific cell classes without having to know the cell classes.

Usage

gecd_CellDistinguisher(exprLinear, genesymb = NULL, numCellClasses = 2,
  minDistinguisherAlternatives = 1, maxDistinguisherAlternatives = 5,
  minAlternativesLengthsNormalized = 0.7, expressionQuantileForScale = 0.75,
  expressionQuantileForFilter = 1, expressionConcentrationRatio = 0.333,
  probesWithGenesOnly = FALSE, verbose = 0)

Arguments

exprLinear	exprLinear is a matrix of (linear / non-logarithmic) expression values that has one row for each marker (a.k.a. probe or gene) and one column for each sample.
genesymb	genesymb is the associated gene names for the rows of exprLinear. They are not passed in as names(exprLinear) because multiple rows (probes) may map to the same gene.
numCellClasses	numCellClasses is the number of cell classes for which the function will locate class-specific distinguishers. Default is 2.
minDistinguisherAlternatives	minDistinguisherAlternatives is the minimum number of distinguishers to be located for each cell class. Default is 1.
maxDistinguisherAlternatives	maxDistinguisherAlternatives is the maximum number of distinguishers to be located for each cell class. Default is 5.
minAlternativesLengthsNormalized	minAlternativesLengthsNormalized is the threshold for bestLengthsNormalized values. Only alternatives with a value at least as large as the threshold are retained. Default is 0.7. 0.0 is no filtering.
expressionQuantileForScale	expressionQuantileForScale is in [0, 1]. Low values (e.g., 0.40) favor distinguishers for a cell class that have very little expression in the other cell classes. High values (e.g., 0.80) shift the balance towards distinguishers with high expression values. Default is 0.75. 0.0 is no preference for highly expressed genes.
expressionQuantileForFilter	Default is 1.0. 1.0 is no filtering. 0.999 is some filtering.
expressionConcentrationRatio	Default is 0.333. 0.0 is no filtering.
probesWithGenesOnly	Default is FALSE.
verbose	verbose indicates amount of printed output. Default is 0.

Examples

See https://github.com/GeneralElectric/CellDistinguisher/blob/master/README.txt

GeneralElectric/CellDistinguisher documentation built on May 18, 2020, 6:34 a.m.