readTCGA: Read TCGA data to the tidy format

In RTCGA: The Cancer Genome Atlas Data Integration

Description

readTCGA function allows to read unzipped files:

• clinical data - Merge_Clinical.Level_1

• rnaseq data (genes' expressions) - rnaseqv2__illuminahiseq_rnaseqv2

• genes' mutations data - Mutation_Packager_Calls.Level

• Reverse phase protein array data (RPPA) - protein_normalization__data.Level_3

• Merge transcriptome agilent data (mRNA) - Merge_transcriptome__agilentg4502a_07_3__unc_edu__Level_3__unc_lowess_normalization_gene_level__data.Level_3

• miRNASeq data - Merge_mirnaseq__illuminaga_mirnaseq__bcgsc_ca__Level_3__miR_gene_expression__data.Level_3 or "Merge_mirnaseq__illuminahiseq_mirnaseq__bcgsc_ca__Level_3__miR_gene_expression__data.Level_3"

• methylation data - Merge_methylation__humanmethylation27

• isoforms data - Merge_rnaseqv2__illuminahiseq_rnaseqv2__unc_edu__Level_3__RSEM_isoforms_normalized__data.Level_3

from TCGA project. Those files can be easily downloded with downloadTCGA function. See examples.

Usage

readTCGA(path, dataType, ...)

Arguments

path	See details and examples.
dataType	One of 'clinical', 'rnaseq', 'mutations', 'RPPA', 'mRNA', 'miRNASeq', 'methylation', 'isoforms' depending on which type of data user is trying to read in the tidy format.
...	Further arguments passed to the as.data.frame.

Details

All cohort names can be checked using: sub( x = names( infoTCGA() ), '-counts', '').

Parameter path specification:

• If dataType = 'clinical' a path to a cancerType.clin.merged.txt file.

• If dataType = 'mutations' a path to the unzziped folder Mutation_Packager_Calls.Level containing .maf files.

• If dataType = 'rnaseq' a path to the uzziped file rnaseqv2__illuminahiseq_rnaseqv2__unc_edu__Level_3__RSEM_genes_normalized__data.Level.

• If dataType = 'RPPA' a path to the unzipped file in folder protein_normalization__data.Level_3.

• If dataType = 'mRNA' a path to the unzipped file cancerType.transcriptome__agilentg4502a_07_3__unc_edu__Level_3__unc_lowess_normalization_gene_level__data.data.txt.

• If dataType = 'miRNASeq' a path to unzipped files cancerType.mirnaseq__illuminahiseq_mirnaseq__bcgsc_ca__Level_3__miR_gene_expression__data.data.txt or cancerType.mirnaseq__illuminaga_mirnaseq__bcgsc_ca__Level_3__miR_gene_expression__data.data.txt

• If dataType = 'methylation' a path to unzipped files cancerType.methylation__humanmethylation27__jhu_usc_edu__Level_3__within_bioassay_data_set_function__data.data.txt.

• If dataType = 'isoforms' a path to unzipped files cancerType.rnaseqv2__illuminahiseq_rnaseqv2__unc_edu__Level_3__RSEM_isoforms_normalized__data.data.txt.

Value

An output:

• If dataType = 'clinical' a data.frame with clinical data.

• If dataType = 'rnaseq' a data.frame with rnaseq data.

• If dataType = 'mutations' a data.frame with mutations data.

• If dataType = 'RPPA' a data.frame with RPPA data.

• If dataType = 'mRNA' a data.frame with mRNA data.

• If dataType = 'miRNASeq' a data.frame with miRNASeq data.

• If dataType = 'methylation' a data.frame with methylation data.

• If dataType = 'isoforms' a data.frame with isoforms data.

Issues

If you have any problems, issues or think that something is missing or is not clear please post an issue on https://github.com/RTCGA/RTCGA/issues.

Author(s)

Marcin Kosinski, [e-mail]

Witold Chodor, [e-mail]

See Also

RTCGA website http://rtcga.github.io/RTCGA/Download.html.

Other RTCGA: RTCGA-package, boxplotTCGA, checkTCGA, convertTCGA, datasetsTCGA, downloadTCGA, expressionsTCGA, heatmapTCGA, infoTCGA, installTCGA, kmTCGA, mutationsTCGA, pcaTCGA, survivalTCGA, theme_RTCGA

Examples

## Not run:  

##############
##### clinical
##############

dir.create('data')

# downloading clinical data
# dataset = "clinical" is default parameter so we may omit it
downloadTCGA( cancerTypes = c('BRCA', 'OV'),
              destDir = 'data' )

    
# reading datasets    
sapply( c('BRCA', 'OV'), function( element ){
    folder <- grep( paste0( '(_',element,'\\.', '|','_',element,'-FFPE)', '.*Clinical'),
                    list.files('data/'),value = TRUE)
    path <- paste0( 'data/', folder, '/', element, '.clin.merged.txt')
    assign( value = readTCGA( path, 'clinical' ), 
            x = paste0(element, '.clin.data'), envir = .GlobalEnv)
    })
     
############
##### rnaseq
############

dir.create('data2')

# downloading rnaseq data
downloadTCGA( cancerTypes = 'BRCA', 
dataSet = 'rnaseqv2__illuminahiseq_rnaseqv2__unc_edu__Level_3__RSEM_genes_normalized__data.Level',
              destDir = 'data2' )

# shortening paths and directories
list.files( 'data2/') %>% 
    file.path( 'data2', .) %>%
    file.rename( to = substr(.,start=1,stop=50))

# reading data
list.files( 'data2/') %>% 
    file.path( 'data2', .) -> folder

folder %>%
    list.files %>%
    file.path( folder, .) %>%
    grep( pattern = 'illuminahiseq', x = ., value = TRUE) -> pathRNA
readTCGA( path = pathRNA, dataType = 'rnaseq' ) -> my_data


###############
##### mutations
###############

# Example directory in which untarred data will be stored
dir.create('data3')


downloadTCGA( cancerTypes = 'OV', 
              dataSet = 'Mutation_Packager_Calls.Level',
              destDir = 'data3' )

# reading data
list.files( 'data3/') %>% 
    file.path( 'data3', .) -> folder

readTCGA(folder, 'mutations') -> mut_file

#################
##### methylation
#################

# Example directory in which untarred data will be stored
dir.create('data4')

# Download KIRP methylation data and store it in data4 folder
cancerType = "KIRP"
downloadTCGA(cancerTypes = cancerType,
             dataSet = "Merge_methylation__humanmethylation27",
             destDir = "data4")

# Shorten path of subdirectory with KIRP methylation data
list.files(path = "data4", full.names = TRUE) %>%
    file.rename(from = ., to = file.path("data4", paste0(cancerType, ".methylation")))

# Remove manifest.txt file
list.files(path = "data4", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE) %>% 
    grep("MANIFEST.txt", x = ., value = TRUE) %>%
    file.remove()

# Read KIRP methylation data
path <- list.files(path = "data4", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE)

KIRP.methylation <- readTCGA(path, dataType = "methylation")


##########
##### RPPA
##########

# Directory in which untarred data will be stored
dir.create('data5')

# Download BRCA RPPA data and store it in data5 folder
cancerType = "BRCA"
downloadTCGA(cancerTypes = cancerType,
             dataSet = "protein_normalization__data.Level_3",
             destDir = "data5")

# Shorten path of subdirectory with BRCA RPPA data
list.files(path = "data5", full.names = TRUE) %>%
    file.rename(from = ., to = file.path("data5", paste0(cancerType, ".RPPA")))

# Remove manifest.txt file
list.files(path = "data5", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE) %>% 
    grep("MANIFEST.txt", x = ., value = TRUE) %>%
    file.remove()

# Read BRCA RPPA data
path <- list.files(path = "data5", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE)

BRCA.RPPA <- readTCGA(path, dataType = "RPPA")


##########
##### mRNA
##########

# Directory in which untarred data will be stored
dir.create('data6')

# Download UCEC mRNA data and store it in data6 folder
cancerType = "UCEC"
downloadTCGA(cancerTypes = cancerType,
dataSet = "Merge_transcriptome__agilentg4502a_07_3__unc_edu__Level_3__unc_lowess_normalization_gene_level__data.Level_3",
             destDir = "data6")

# Shorten path of subdirectory with UCEC mRNA data
list.files(path = "data6", full.names = TRUE) %>%
    file.rename(from = ., to = file.path("data6",paste0(cancerType, ".mRNA")))

# Remove manifest.txt file
list.files(path = "data6", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE) %>% 
    grep("MANIFEST.txt", x = ., value = TRUE) %>%
    file.remove()

# Read UCEC mRNA data
path <- list.files(path = "data6", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE)

UCEC.mRNA <- readTCGA(path, dataType = "mRNA")

##############
##### miRNASeq
##############

# Directory in which untarred data will be stored
dir.create('data7')

# Download BRCA miRNASeq data and store it in data7 folder
# Remember that miRNASeq data are produced by two machines:
# Illumina Genome Analyzer and Illumina HiSeq 2000 machines
cancerType <- "BRCA"
downloadTCGA(cancerTypes = cancerType,
dataSet = "Merge_mirnaseq__illuminaga_mirnaseq__bcgsc_ca__Level_3__miR_gene_expression__data.Level_3",
             destDir = "data7")

downloadTCGA(cancerTypes = cancerType,
dataSet = "Merge_mirnaseq__illuminahiseq_mirnaseq__bcgsc_ca__Level_3__miR_gene_expression__data.Level_3",
             destDir = "data7")

# Shorten path of subdirectory with BRCA miRNASeq data
list.files(path = "data7", full.names = TRUE) %>%
    sapply(function(path){
        if (grepl(pattern = "illuminaga", path)){
            file.rename(from = grep(pattern = "illuminaga", path, value = TRUE),
                        to = file.path("data7",paste0(cancerType, ".miRNASeq.illuminaga")))
        } else if (grepl(pattern = "illuminahiseq", path)){
            file.rename(from = grep(pattern = "illuminahiseq", path, value = TRUE),
                        to = file.path("data7",paste0(cancerType, ".miRNASeq.illuminahiseq")))
        }
    })
    
# Remove manifest.txt file
list.files(path = "data7", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE) %>% 
    grep("MANIFEST.txt", x = ., value = TRUE) %>%
    file.remove()

# Read BRCA miRNASeq data
path <- list.files(path = "data7", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE)
path_illuminaga <- grep("illuminaga", path, fixed = TRUE, value = TRUE)
path_illuminahiseq <- grep("illuminahiseq", path, fixed = TRUE, value = TRUE)

BRCA.miRNASeq.illuminaga <- readTCGA(path_illuminaga, dataType = "miRNASeq")
BRCA.miRNASeq.illuminahiseq <- readTCGA(path_illuminahiseq, dataType = "miRNASeq")

BRCA.miRNASeq.illuminaga <- cbind(machine = "Illumina Genome Analyzer", BRCA.miRNASeq.illuminaga)
BRCA.miRNASeq.illuminahiseq <- cbind(machine = "Illumina HiSeq 2000", BRCA.miRNASeq.illuminahiseq)

BRCA.miRNASeq <- rbind(BRCA.miRNASeq.illuminaga, BRCA.miRNASeq.illuminahiseq)

##############
##### isoforms
##############

# Directory in which untarred data will be stored
dir.create('data8')

# Download ACC isoforms data and store it in data8 folder
cancerType = "ACC"
downloadTCGA(cancerTypes = cancerType,
dataSet = "Merge_rnaseqv2__illuminahiseq_rnaseqv2__unc_edu__Level_3__RSEM_isoforms_normalized__data.Level_3",
             destDir = "data8")

# Shorten path of subdirectory with ACC isoforms data
list.files(path = "data8", full.names = TRUE) %>%
    file.rename(from = ., to = file.path("data8",paste0(cancerType, ".isoforms")))

# Remove manifest.txt file
list.files(path = "data8", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE) %>% 
    grep("MANIFEST.txt", x = ., value = TRUE) %>%
    file.remove()

# Read ACC isoforms data
path <- list.files(path = "data8", full.names = TRUE) %>%
    list.files(path = ., full.names = TRUE)

ACC.isoforms <- readTCGA(path, dataType = "isoforms")


## End(Not run)

RTCGA documentation built on Nov. 8, 2020, 5:11 p.m.