curatedTCGAData-helpers: Helper functions for managing MultiAssayExperiment from...

In waldronlab/TCGAmisc: TCGA utility functions for data management

Helper functions for managing MultiAssayExperiment from curatedTCGAData

Description

Additional helper functions for cleaning and uncovering metadata within a downloaded MultiAssayExperiment from curatedTCGAData.

Usage

getSubtypeMap(multiassayexperiment)

getClinicalNames(diseaseCode)

TCGAsplitAssays(multiassayexperiment, sampleCodes = NULL, exclusive = FALSE)

sampleTables(multiassayexperiment, vial = FALSE)

Arguments

multiassayexperiment	A MultiAssayExperiment object
diseaseCode	A TCGA cancer code (e.g., "BRCA")
sampleCodes	character (default NULL) A string of sample type codes (refer to data(sampleTypes); TCGAsplitAssays section)
exclusive	logical (default FALSE) Whether to return only assays that contain all codes in sampleCodes
vial	(logical default FALSE) whether to display vials in the table output

Details

Note that for getSubtypeMap, the column of in-data variable names may need to go through make.names to be found in the colData of the MultiAssayExperiment.

Value

• getSubtypeMap: A data.frame with explanatory names and their in-data variable names. They may not be present for all cancer types.

• getClinicalNames: A vector of common variable names that may be found across several cancer disease codes.

getSubtypeMap

provides a two column data.frame with interpreted names and in-data variable names. 'Name' usually refers to the colData row names a.k.a. the patientID.

getClinicalNames

provides a vector of common variable names that exist in the colData DataFrame of a curatedTCGAData MultiAssayExperiment object. These variables are directly obtained from the BroadFirehose clinical data (downloaded with getFirehoseData) and tend to be present across cancer disease codes.

TCGAsplitAssays

Separates samples by indicated sample codes into different assays in a MultiAssayExperiment. Refer to the sampleTypes data object for a list of available codes. This operation generates n times the number of assays based on the number of sample codes entered. By default, all assays will be split by samples present in the data.

sampleTables

Display all the available samples in each of the assays

Examples

library(curatedTCGAData)

gbm <- curatedTCGAData("GBM", c("RPPA*", "CNA*"), version = "2.0.1", FALSE)

getSubtypeMap(gbm)

sampleTables(gbm)

TCGAsplitAssays(gbm, c("01", "10"))

getClinicalNames("COAD")

waldronlab/TCGAmisc documentation built on Dec. 19, 2024, 2:10 p.m.