This is a test version of Biostars. For the public version, visit https://www.biostars.org.
vcf2bed giving empty bed file as output

I'm trying to use vcf2bed to convert a ~0.5TB .vcf file to .bed and am unable to figure out why this command isn't working. I get a blank file as the output. The job completes in a second and gives me nothing but a blank file. Please let me know what might be the problem here.

vcf2bed < /path/input.vcf > /path/output.bed

Also, it would be nice if anybody could give me an estimate on how large the output .bed file will be when working with a 0.5TB input.

To be sure the format of the input is correct, Here's part of the file: Image of several dozen rows of .vcf file after metadata

Is my .vcf input in the right format? It appears to be the same as what is required in the BEDOPS manual (https://media.readthedocs.org/pdf/bedops/latest/bedops.pdf)

My input command and .vcf appeaer to be consistent with instructions in the notebook and several forum posts, so I don't know how to get past this. Why is the .bed output 0 bytes?

Thanks for the help.

gwas genome genomics
• 6,735 views • link
updated by aweaver7204 • aweaver7204 • written by michael.nagle michael.nagle

Wild guess: wrong path to input file. What's your exact command line?

• 0 views • link
written by jomo018 jomo018

Did this solution work for you because I am having the same issue and this solution did not fix it. Any help would be greatly appreciated.

• 0 views • link
written by aweaver7204 • aweaver7204

1 answer

It is likely that your /tmp folder is filling up with intermediate data during the sorting step. Some /tmp or swap folders are not large enough to hold intermediate results.

Use --sort-tmpdir <dir> with vcf2bed to specify an alternative directory <dir> that can contain more than 500 GB of data (a worst-case scenario, where all variants are on one chromosome).

Alternatively, use --do-not-sort with vcf2bed to keep the result unsorted, and then sort afterwards with sort-bed --tmpdir <dir>, which accomplishes the same result.

If the BED file is too large, you can use vcf2starch to create a Starch archive from the BED file. This will be about twice as efficient as compression with gzip. The BEDOPS documentation describes Starch files and the format in more detail. BEDOPS tools work natively with Starch as well as BED.

• 0 views • link
written by Alex Reynolds Alex Reynolds

Log in to answer this question.

More posts like this