A curated catalogue of human genomic structural variation




Variant Details

Variant: essv9486



Internal ID9631624
Landmark
Location Information
TypeCoordinatesAssemblyOther Links
Innerchr20:30258517..30305953hg38UCSC Ensembl
Outerchr20:30258517..30396019hg38UCSC Ensembl
Innerchr20:29493193..29540629hg19UCSC Ensembl
Outerchr20:29493193..29630695hg19UCSC Ensembl
Innerchr20:28106854..28154290hg18UCSC Ensembl
Outerchr20:28106854..28244356hg18UCSC Ensembl
Innerchr20:28106854..28154290hg17UCSC Ensembl
Outerchr20:28106854..28244356hg17UCSC Ensembl
Cytoband20q11.1
Allele length
AssemblyAllele length
hg38137503
hg19137503
hg18137503
hg17137503
Variant TypeCNV loss
Copy Number
Allele State
Allele Origin
Probe Count
Validation Flag
Merged StatusS
Merged Variantsesv2756736
Supporting Variants
SamplesNA19154
Known GenesFRG1B
MethodSNP array
AnalysisThe algorithm used to call CNVs using the 500K EA platform was developed to accurately define CNV regions using a large set of reference samples and is described in detail in a separate publication (Komura 2006). The algorithm contains three major parts: 1) Intensity pre-processing using an improved version of Genomic Imbalance Map (GIM) (Ishikawa et al. 2005), including probe selection, noise reduction, normalization, and intensity ratio adjustment based on affinity differences between alleles of a SNP, 2) CNV extraction, which identifies CNVs from all pair-wise comparisons using a modified SW-ARRAY, and 3) A copy number inference step which utilizes signal ratios and SNP information to more precisely define CNV boundaries and the copy number within each region.
PlatformAffymetrix GeneChip Early Access Mapping 500K Set Array (250K_Nsp_SNP)
Comments
ReferenceRedon_et_al_2006
Pubmed ID17122850
Accession Number(s)essv9486
Frequency
Sample Size270
Observed Gain0
Observed Loss1
Observed Complex0
Frequencyn/a


Hosted by The Centre for Applied Genomics
Grant support for DGV
Please read the usage disclaimer