get_diffexed_genes: get_diffexed_genes

Description Usage Arguments Value

Description

Extract genes from a limma-fit object (resulting from an eBayes call) that are differentially expressed. Two different methods are used for calling differential expression: any_probe - at least one probe for that gene must be differentially expressed. all_probes - every probe for that gene must be differentially expressed (and must be differentially expressed in the same direction). This doesn't return information about the direction of change, and only returns a vector of gene symbols or ids, as extracted from the genes dataframe in the limma-fit object fit.

Usage

1
2
get_diffexed_genes(fit, gene_column = "symbol", diffex_type = c("any_probe",
  "all_probes"), ...)

Arguments

fit

A limma fit object (class: MArrayLM) as returned by eBayes. If the initial model/analysis contained more than one contrast the user should select a single contrast prior to passing fit into get_diffexed_genes, by doing fit[, some.column] - if info on more than one contrast is present in the fit object, the function throws an error.

gene_column

The name of the column within fit$genes that contains the gene-symbols or gene-ids that are to be returned by this function. For example, if you want to receive a subset of the "entrez.id" column of fit$genes then specify gene_column = "entrez.id". The function dies if the specified column is absent from the fit$genes dataframe.

diffex_type

The user selects either "any_probe" or "all_probes" to indicate the type of analysis to be performed.

...

Further arguments that are to be passed through to limma::decideTests. For example, if you want to set the lfc filter to 1 (in decideTests; corresponding to keeping probes only if they are at least 2-fold up/down-regulated), you can specify it as follows: get_diffexed_genes(some.fit, some.column, some.type, lfc = 1).

Value

A subvector of the inidicated 'gene_column' in fit fit$genes annotation dataframe.


russHyde/ee.marray.functions documentation built on May 6, 2019, 7:32 a.m.