Error in the gdcRNAMerge()

[epfarias]

I've downloaded all the data correctly and when i get to the merge moment i recive this error

####### Merge RNAseq data #######
rnaCounts <- gdcRNAMerge(metadata = metaMatrix.RNA,
path = rnadir, # the folder in which the data stored
organized = FALSE, # if the data are in separate folders
data.type = 'RNAseq')

############### Merging RNAseq data ################

This step may take a few minutes

Error in scan(file = file, what = what, sep = sep, quote = quote, dec = dec, : line 2 did not have 9 elements

Anyone could help me with this problem ?

[LGROOVE]

Hi I have the same persistent problem!

[epfarias]

Hi @LGROOVE i've soved this issue downloading the data from Xenabrowser and adapting the informations from dataset to the metamatrix_MIR

[LGROOVE]

Hi there, did you modify the gene expression tsv files?

[An-05]

I have the same problem. How was the problem solved?

[pegasusCN]

TCGA has changed its data format, so original function gdcRNAMerge() won't work. Try this:

myrnaMerge <- function (metadata, path, data.type, organized = FALSE)
{
if (organized == TRUE) {
filenames <- file.path(path, metadata$file_name, fsep = .Platform$file.sep)
}
else {
filenames <- file.path(path, metadata$file_id, metadata$file_name,
fsep = .Platform$file.sep)
}
if (data.type == "RNAseq") {
message("############### Merging RNAseq data ################\n",
"### This step may take a few minutes ###\n")
rnaMatrix <- do.call("cbind", lapply(filenames,
function(fl) read.table(gzfile(fl), skip = 6)$V4))
rownames(rnaMatrix) <- read.table(gzfile(filenames[1]), skip = 6)$V1
rownames(rnaMatrix) <- unlist(lapply(strsplit(rownames(rnaMatrix),
".", fixed = TRUE), function(gene) gene[1]))
colnames(rnaMatrix) <- metadata$sample
# rnaMatrix <- rnaMatrix[biotype$ensemblID, ]
nSamples = ncol(rnaMatrix)
nGenes = nrow(rnaMatrix)
message(paste("Number of samples: ", nSamples, "\n",
sep = ""), paste("Number of genes: ", nGenes, "\n",
sep = ""))
return(rnaMatrix)
}
else if (data.type == "pre-miRNAs") {
message("############### Merging pre-miRNAs data ################\n",
"### This step may take a few minutes ###\n")
rnaMatrix <- do.call("cbind", lapply(filenames, function(fl) read.delim(fl)$read_count))
rownames(rnaMatrix) <- read.delim(filenames[1])$miRNA_ID
colnames(rnaMatrix) <- metadata$sample
nSamples = ncol(rnaMatrix)
nGenes = nrow(rnaMatrix)
message(paste("Number of samples: ", nSamples, "\n",
sep = ""), paste("Number of genes: ", nGenes, "\n",
sep = ""))
return(rnaMatrix)
}
else if (data.type == "miRNAs") {
message("############### Merging miRNAs data ###############\n")
mirMatrix <- lapply(filenames, function(fl) cleanMirFun(fl))
mirs <- rownames(mirbase)
mirMatrix <- do.call("cbind", lapply(mirMatrix, function(expr) expr[mirs]))
rownames(mirMatrix) <- mirbase$v21[match(mirs, rownames(mirbase))]
colnames(mirMatrix) <- metadata$sample
mirMatrix[is.na(mirMatrix)] <- 0
nSamples = ncol(mirMatrix)
nGenes = nrow(mirMatrix)
message(paste("Number of samples: ", nSamples, "\n",
sep = ""), paste("Number of miRNAs: ", nGenes, "\n",
sep = ""))
return(mirMatrix)
}
else {
return("error !!!")
}
}

then use this myrnaMerge() to process your data.