Meldrum describes the ability to incorporate CNVs in aCGH as the next wave. In the same way that there is normal variation at the level of individual base pairs, or SNPs, that does not cause phenotypic changes, the growing recognition that the baseline of normal CNV may be greater than originally proposed has sparked efforts to characterize the extent of CNVs. Agilent plans to add CNV information to its aCGH products through refined probe design and new bioinformatics tools that will define the location and relevance of CNVs.
A collaborative study emanating from the Sanger Institute, for example, explored global variation in copy number in the human genome using both array-based CGH and SNP genotyping, describing it as “functionally significant” and identifying 1,447 copy number variable regions (CNVRs) covering 360 megabases, or 12% of the genome. Richard Redon, et.al., noted that “the CNVRs encompassed more nucleotide content per genome than SNPs, underscoring the importance of CNV in genetic diversity and evolution.”
“Array CGH is a powerful addition to the repertoire of clinical cytogenetic methodologies but must be interpreted within the context of traditional cytogenetics,” remarks Mansoor Mohammed, Ph.D., CEO and president of CombiMatrix Molecular Diagnostics (CMDX; www.cmdiagnostics.com).
When the results of aCGH reveal an extra copy of a genomic locus or of an entire chromosome, for example, it is important to understand whether the extra copy is due to a tandem duplication or a third copy conjoined with another chromosome, explains Dr. Mohammed. These possibilities are associated with different risks of recurrence (in future pregnancies) and may point to different etiologies of the chromosomal aberration. In many cases, traditional cytogenetics, guided by the findings of aCGH, are needed to determine if one parent carries a particular balanced translocation that manifests as an imbalance in the chromosome complement of the child.
Fluorescence in situ hybridization (FISH) is the gold standard used to verify the accuracy of aCGH. FISH relies on BACs and other large insert clones. While oligo-based aCGH offers the potential advantage of generating a higher-resolution analysis, it is not yet ready for use in a clinical setting because oligos are incompatible with standard FISH methods, Dr. Mohammed cautions.
“When you get a ratio from an oligo array suggesting a genome copy number aberration, it is difficult to confirm such aberrations within a chromosomal context,” he says. Furthermore, he points out that the higher resolution of oligo aCGH and its potential to detect smaller CNVs throughout the genome could inundate diagnosticians with large amounts of uninterpretable data until the background work has been done to define the scope of normal CNV in various populations.
A sampling of the scientific literature reveals a growing knowledge base supporting the belief that CNVs, whether representing normal variation or linked to disease predisposition, are more common than previously recognized. Additionally, the number of papers on the application of aCGH in studies of model organisms continues to expand.
A report in Genome Research (2007;17(3):337-347) describes the use of aCGH to screen for novel induced deletions in C. elegans and to determine the amount of natural gene content variation between two strains of C. elegans, shown to be almost 2%. The technology enabled detection of both large (50 kb) multigene deletions and of small (1 kb) single-gene deletions in homozygous and heterozygous samples.
Timothy Graubert, M.D., and colleagues from Washington University in St. Louis along with NimbleGen Systems (www.nimblegen.com) generated a high-resolution map of segmental DNA CNV in the mouse genome, comparing 21 inbred mouse strains using aCGH. They identified CNVs ranging in size from 21 to 2,002 kb, many associated with known polymorphic traits, and were able to use copy number information to predict the phenotype of previously uncharacterized mouse strains.