Structural Variants Standard Dataset

Question

[PoGibas]

I am testing structural variants (deletions, duplications, inversions, insertions) calling tools and need a standard dataset to validate my calls.

Where can I find benchmark bam and corresponding vcf file for specific individual?

1000Genomes has calls for all the SV types only in the pilot release. But I am not able to find out which bam file (WGS or WES) should I use to call SV's and which vcf file should I use to check my calls.

Any help with an example will be appreciated.

Answer 1

[Giovanni M Dall'Olio]

This is the folder containing the variants called in 1000 Genomes: ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20110521/

If you look at the README in that folder, you will find a description of which files have been used to call these variants, and a short description of the filters applied.

Note that in these folders, Structural and Single Base Variants are merged together, in the same files. To extract the structural variants, you can use the option --keep-indels from the latest version of vcftools ( http://vcftools.sourceforge.net/options.html )

[PoGibas]

And bam files should be downloaded from ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/phase1/data/ ? If so, why bam file size differ (for example: low coverage Illumina bam file size varies from 9 to 24Gb)?

Answer 2

[Dan]

You could try generating simulated SVs then simulated reads from them? It's not a very complete list, but some NGS simulators are listed on SEQwiki: http://seqanswers.com/wiki/Special:BrowseData/Bioinformatics_application?Bioinformatics_method=Simulation