Regarding Human Exon Variant Reference

Question

[k.manojkumar2991]

Hi,

I'm new to NGS data analysis. I have downloaded the Whole human Exome data from the SRA (SRA ID SRR098401) and I started to align them using BWA and I have called the variants using Gatk. now I want to compare the variants to check my programs are working fine or not . my bad luck I cant able to find the Reference variants for the whole exome. can any one tell me where can I find the Reference Variants for the sample I have taken from the SRA for comparing my alignment program with the variant calling is working fine or not

Answer 1

[Garan]

You'd probably be better off using the Genome in a Bottle reference data, which has a set of high confidence variant calls obtained from various sequencing technologies (https://sites.stanford.edu/abms/giab) if you're trying to validate your pipeline.

For NA12878

Fastqs (high coverage)

ftp://ftp-trace.ncbi.nlm.nih.gov/giab/data/NA12878/NIST_NA12878_HG001_HiSeq_300x

High confidence variant calls

ftp://ftp-trace.ncbi.nlm.nih.gov/giab/ftp/release/NA12878_HG001/

Illumina also provide data for NA12878 (not sure if this was integrated into the GIAB high confidence calls), especially useful since they have the full CEPH pedigree 1463 (inc. NA12878)

http://www.illumina.com/platinumgenomes/

If you're only interested in a set of intervals specific for a certain Exome panel, just run something like GATK SelectVariants with the intervals on the results VCFs.

[k.manojkumar2991]

Thanks @Garan But I need only the exome from the NIST but it contains whole genome.

Answer 2

[Amitm]

Hi,

I guess what you mean are the calls that were obtained by the 1000G consortia because the sample you are analyzing belongs there. The page here tells more where you can find the "reference" calls in VCF format - http://www.1000genomes.org/data