converted TCGA file to bed format but it doesn't want to be read in R

[SaraYones]

I used a tool to download TCGA data and convert it into BED format it is called TCGA2BED https://bmcbioinformatics.biomedcent...859-016-1419-5

when i open the bed file it looks like this

chr1 61735 2448306 * 486 0.0678 N
chr1 2449711 2501338 * 22 0.5571 N
chr1 2503034 2959610 * 142 0.0025 N
chr1 2960410 3058370 * 32 0.3827 N
chr1 3068646 21751244 * 11226 -0.0042 N

the data type of this file is CNV so I am trying now to map genome regions to gene symbols in CNV files . I downloaded the RefSeq gene coordinates in bed format from USCS. i am now trying to use R (bedr package) to intersect the regions. I am using function import in rtracklayer to read the files

a=import("/Users/sarayones/Downloads/TCGA_LAML/CNV/CNVoutputBed/MyFile.bed", format="bed")

When I try to read the converted TCGA bed file (in the above format) using import it gives me this error
Error in scan(file = file, what = what, sep = sep, quote = quote, dec = dec, :
scan() expected 'an integer', got 'N'

I validated that the converted file is correct using bed-ops
bedops --ec --everything "/Users/sarayones/Downloads/TCGA_LAML/CNV/CNVoutputBed/MyFile.bed" 1> /dev/null

and it didn't give any error but I am kind of doubtful that there is something wrong with the format. Any advices?

[neavemj]

Hi Sara,

That error seems to suggest that the 7th column of your BED file should be a number (i.e. an integer) rather than the letter "N". Maybe you could remove the Ns and try again?

I also thought that one of those columns should indicate the strand (+ or -), but I can't see that in your file. So there might be some other things wrong with the conversion..

Good luck!

Matt.