Discovery of structural variation in the human genome
The advent of genome-scanning technologies and comparative DNA analysis has uncovered a significant extent of ‘structural variation’ in the human genome. Structural variants can include microscopic and more commonly submicroscopic deletions, duplications, and large-scale copy number variants - collectively termed copy number variants (CNVs) or polymorphisms - as well as insertions, inversions and translocations. Rapidly accumulating evidence indicates that structural variants can comprise millions of nucleotides of heterogeneity, having an important contribution to human diversity and disease susceptibility. To generate a CNV map of the human genome our Consortium (with N. Carter, M. Hurles, C. Tyler-Smith: Wellcome Trust Sanger Institute; C. Lee: Brigham Women's Hospital/Harvard; K Jones: Affymetrix; H. Aburatani: Tokyo) screened all 270 individuals from the four HapMap populations with ancestry in Europe, Africa and East Asia for CNVs using two complementary technologies: the Affymetrix 500k-SNP platform and CGH on a microarray containing ~27,000 clones representing the genome tile-path. We rigorously assessed the rate of false positive CNV calls by several independent means, including the validation of hundreds of loci by quantitative PCR and FISH. We identified over a thousand CNVs in these four populations, many of which have not previously been identified, and which contain hundreds of genes and non-coding functional sequences. In a second approach we have performed comparative DNA analysis of human-human and human-chimp assemblies and experimentation to find hundreds of other putative structural variants in the human genome. Our collective data, integrated with all other information, is released in the 'Database of Genomic Variants' (http://projects.tcag.ca/variation/). The database serves as a resource to assist numerous clinical research studies including assessing CNV contribution in the etiology of autism spectrum and other developmental disorders.