Too short deletion in SV from 1000 Genomes

Question

[one]

Hi everybody, I downloaded the variants from 1000 Genomes project phase 3 (e.g ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/ALL.chr14.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz)

When I check structural variants with type = 'DEL' (VT=SV, SVTYPE=DEL), I see mostly reference with one base and alternative as <cn0>. I expect SV would have a much larger size, otherwise, they would be considered as INDELs. (most INDELs in the data are a few bases)

I am new to SVs. Could anyone give some insights ?

Thanks,

Answer 1

[Pierre Lindenbaum]

there are some END/ CIEND/SVLEN... fields in the INFO column .

##INFO=<ID=CIEND,Number=2,Type=Integer,Description="Confidence interval around END for imprecise variants">
##INFO=<ID=END,Number=1,Type=Integer,Description="End coordinate of this variant">
##INFO=<ID=SVLEN,Number=.,Type=Integer,Description="SV length. It is only calculated for structural variation MEIs. For other types of SVs; one may calculate the SV length by INFO:END-START+1, or by finding the difference between lengthes of REF and ALT alleles">
(...)
14  19076395    T   <CN0>   AC=1;AF=0.000199681;AN=5008;CIEND=-17,17;CIPOS=-17,17;CS=DEL_union;END=19079192;NS=2504;SVTYPE=DEL;DP=32427;EAS_AF=0;AMR_AF=0;AFR_AF=0.0008;EUR_AF=0;SAS_AF=0;VT=SV