Entering edit mode
michaelwxf2012
•
0
@michaelwxf2012-14463
Last seen 7.0 years ago
I use package TCGAbiolinks to get access to the DNA methylation data from TCGA. The problem occurs at the prepare part. I have download the data successfully.
#Load Package
library("TCGAbiolinks")
#TCGA query
LUAD.met<-GDCquery(project = "TCGA-LUAD",
legacy = TRUE,
data.category = "DNA methylation",
platform = "Illumina Human Methylation 450")
#Download
GDCdownload(LUAD.met, method = "api", files.per.chunk = 5)
#Prepare
LUAD.methy<-GDCprepare(LUAD.met)
The error shows:
Error: cannot allocate vector of size 3.7 Mb.
Traceback() as followed,
8: copy(ans[[target]])
7: `[.data.table`(y, x, nomatch = if (all.x) NA else 0, on = by,
allow.cartesian = allow.cartesian)
6: y[x, nomatch = if (all.x) NA else 0, on = by, allow.cartesian = allow.cartesian]
5: merge.data.table(df, data, by = "Composite.Element.REF")
4: merge(df, data, by = "Composite.Element.REF")
3: merge(df, data, by = "Composite.Element.REF")
2: readDNAmethylation(files, query$results[[1]]$cases, summarizedExperiment,
unique(query$platform))
1: GDCprepare(LUAD.met)
sesseionInfo() as followed,
R version 3.4.2 (2017-09-28) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows >= 8 x64 (build 9200) Matrix products: default locale: [1] LC_COLLATE=Chinese (Simplified)_China.936 LC_CTYPE=Chinese (Simplified)_China.936 [3] LC_MONETARY=Chinese (Simplified)_China.936 LC_NUMERIC=C [5] LC_TIME=Chinese (Simplified)_China.936 attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] nycflights13_0.2.2 dplyr_0.7.4 bindrcpp_0.2 TCGAbiolinks_2.6.1
I couldnt understand the error meaning if my downloading files may have some errors since there are a large number of files. I hope some professionals can help me dealing with it. Thanks.
Michael Wang
Hi Marcel,
Thank you for your answering. How can I fix this problem? Delete the data and download again?
Yours,
Michael
How much memory (RAM) do you have. The problem as Marcel pointed out is that R was not able to read the data to memory.
You would have to read in chunks then try to bind the objects.
Something like:
#Load Package library("TCGAbiolinks") #TCGA query LUAD.met<-GDCquery(project = "TCGA-LUAD", legacy = TRUE, data.category = "DNA methylation", platform = "Illumina Human Methylation 450") #Download GDCdownload(LUAD.met, method = "api", files.per.chunk = 5) samples <- getResults(LUAD.met)$cases nsamples <- length(getResults(LUAD.met)$cases) step <- 50 for(start in seq(1,nsamples,step)) { end <- start + step if(end > nsamples) end <- nsamples LUAD.met<-GDCquery(project = "TCGA-LUAD", legacy = TRUE, barcode = samples[start:end], data.category = "DNA methylation", platform = "Illumina Human Methylation 450") GDCdownload(LUAD.met, method = "api", files.per.chunk = 5) LUAD.methy<-GDCprepare(LUAD.met,save = T,save.filename = paste0(start,".rda")) } LUAD.met <- NULL for(i in seq(1,nsamples,step)) { print(i) if(is.null(LUAD.met)) { LUAD.met <- get(load(paste0(i,".rda"))) } else { aux <- get(load(paste0(i,".rda"))) aux <- aux[rownames(LUAD.met)] LUAD.met <- SummarizedExperiment::cbind(LUAD.met,aux) } } #PrepareHi dear friend,
My laptop is not professional for bioinformatics with a RAM of 4GB and Core i5 7200U. I have got your meaning and I really appreciate your kind reply. I can understand this script is trying to help download and prepare the cases 50 by 50 so that my laptop can handle.
I will try it as soon as possible and will let you know the outcome at first time. This reply is just to express my sincere appreciation for your warmly help. Thanks very much.
Michael
I'm not sure we will be able to handle all the DNA methylation data with only 4GB. Probably the step in the code that tries to merge the objects will give you the same problem.