How To Obtain At Least 30 Realistlic Example Vcf Files?

Question

[user56]

In our project, we would like to adopt the Variant Call File format (v.4.0 or the newer 4.1). Is there a public source, where we could download a set of 30 or more example VCF files.

I looked at 1000genomes and saw the BAM files, but they are large and I could not find any VCF files in 1000genomes FTP site. Were they created or they are only in the process of doing so?

[sanbisamples]

Use the 1000 Genomes data slicer. It has a nice graphical user interface and easy to understand examples that will help you create small VCF files.

[Sudeep]

is it necessary that all these 30 VCF files are for a single species or could it be from any species ?

Answer 1

[Malachi Griffith]

I believe that the VCF files from 1000 genomes are available here from EBI and here from NCBI. 1000 genomes has a fairly detailed description of the VCF file format.

Answer 2

[user56]

From my own investigation, it seems that in this FTP folder there is a larger set. All populations are in one file. (it should be 1000 genomes)

ftp://ftp-trace.ncbi.nih.gov/1000genomes/ftp/release/20110521/

only chomosome 6 variants have, however, 9 GB.

[Eponymous]

More recent release: ftp://ftp-trace.ncbi.nih.gov/1000genomes/ftp/release/20130502/

Answer 3

[Giovanni M Dall'Olio]

You can use tabix to extract subsets of the vcf files from the 1000genomes websites. Thanks to the fact that tabix uses a index file, you will be able to download only portions of the files, without having to download everything in local.

Example:

tabix -h  ftp://ftp-trace.ncbi.nih.gov/1000genomes/ftp/release/20101123/interim_phase1_release/ALL.chr12.phase1.projectConsensus.genotypes.vcf.gz 1:10000,200000

Have a look at this discussion for more help.

[Timothée Flutre]

As of today, this cmd-line doesn't return anything. Here is a quick fix (use "12" instead of "1" and "-" instead of ","):

tabix -h ftp://ftp-trace.ncbi.nih.gov/1000genomes/ftp/release/20101123/interim_phase1_release/ALL.chr12.phase1.projectConsensus.genotypes.vcf.gz 12:10000-200000

Answer 4

[user56]

I was able to use Complete Genomics public data set of 69 genomoes. Initially they did not have index file for me to use tabix right away, but after a forum post, and few days, they created those so I did not have to download the full 70+GB of data.

Answer 5

[Pierre Lindenbaum]

You could ask Zev Kronenberg for his 200 Danish exomes in VCF.