readData2eset: Create ExpressionSet object from expression data

In frankRuehle/systemsbio: Streamlined Analysis and Integration of Systems Biology Data

Description

ExpressionSet object assembled from expression, feature and phenotype data.

Usage

readData2eset(
  exprsData,
  phenoData,
  featureData = NULL,
  ProbeID = "PROBE_ID",
  sampleColumn = "Sample_Name",
  groupColumn = "Sample_Group",
  experimentData = MIAME(),
  annotation = character(),
  transform = log2
)

Arguments

exprsData	dataframe or character with filepath to expression data to be loaded. If no rownames are included, a column containing ProbeIDs is expected and used as feature names.
phenoData	dataframe or character with filepath to sample sheet / phenotype data to be loaded.
featureData	dataframe or character with filepath to feature data to be loaded. If NULL, non-numeric data columns from 'exprsData' are used as feature data (if any).
ProbeID	character with name of the column in 'exprsData' and 'featureData' that contains identifiers used to uniquely identify each probe.
sampleColumn	character with name of sample column in 'phenoData'.
groupColumn	character with name of group column in 'phenoData'.
experimentData	MIAME object with optional experiment description.
annotation	character string specifying array annotation package (e.g. "Humanv4"). character() if not available.
transform	optional function definition to transform expression data (e.g. log2 for logarithm of base 2). if NULL, no data transformation is performed.

Details

Expression data, phenotype data and optional feature data are used to make an ExpressionSet object. Only numeric columns in exprsData are used. If no feature data is given, non-numeric data columns from exprsData are used as feature data (if any). If dataframes don't contain rownames as identifier, the column ProbeID is used as rownames in exprsData and featureData and column sampleColumn in phenoData. Rownames of exprsData must match with rownames of featureData (if given). Column names of exprsData must match with rownames of phenoData. Characters "/" and "-" in sample names or group names are replaced by ".". If either gene symbols or EntrezIDs are missing in feature data, this annotation is added using the corresponding annotation package if available (species name is derived from annotation). The optional function given in transform is used to transform the expression data.

Value

ExpressionSet object

Author(s)

Frank Ruehle

frankRuehle/systemsbio documentation built on Sept. 14, 2020, 1:18 a.m.